2025-09-09T14:02:53.5121142Z Current runner version: '2.328.0' 2025-09-09T14:02:53.5126563Z Runner name: 'i-0164110f67924097b' 2025-09-09T14:02:53.5127275Z Runner group name: 'default' 2025-09-09T14:02:53.5128071Z Machine name: 'ip-10-0-57-110' 2025-09-09T14:02:53.5130775Z ##[group]GITHUB_TOKEN Permissions 2025-09-09T14:02:53.5133023Z Contents: read 2025-09-09T14:02:53.5133537Z Metadata: read 2025-09-09T14:02:53.5134034Z ##[endgroup] 2025-09-09T14:02:53.5135885Z Secret source: Actions 2025-09-09T14:02:53.5136509Z Prepare workflow directory 2025-09-09T14:02:53.5686432Z Prepare all required actions 2025-09-09T14:02:53.5723014Z Getting action download info 2025-09-09T14:02:53.9455227Z Download action repository 'actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683' (SHA:11bd71901bbe5b1630ceea73d27597364c9af683) 2025-09-09T14:02:54.2898178Z Download action repository 'pytorch/pytorch@main' (SHA:4dd73e659a8fd4872e5f49cfd72e420fa7c4e6c9) 2025-09-09T14:03:09.4448821Z Download action repository 'actions/download-artifact@d3f86a106a0bac45b974a628896c90dbdf5c8093' (SHA:d3f86a106a0bac45b974a628896c90dbdf5c8093) 2025-09-09T14:03:09.8511227Z Download action repository 'pmeier/pytest-results-action@a2c1430e2bddadbad9f49a6f9b879f062c6b19b1' (SHA:a2c1430e2bddadbad9f49a6f9b879f062c6b19b1) 2025-09-09T14:03:10.0017026Z Download action repository 'actions/upload-artifact@ea165f8d65b6e75b540449e92b4886f43607fa02' (SHA:ea165f8d65b6e75b540449e92b4886f43607fa02) 2025-09-09T14:03:10.5529509Z Getting action download info 2025-09-09T14:03:10.7255634Z Uses: pytorch/test-infra/.github/workflows/linux_job_v2.yml@refs/heads/main (e502b6d9079a2a411c68046e8a7694b851c5df33) 2025-09-09T14:03:10.7259721Z ##[group] Inputs 2025-09-09T14:03:10.7261147Z script: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:10.7263405Z timeout: 180 2025-09-09T14:03:10.7263689Z runner: linux.g5.12xlarge.nvidia.gpu 2025-09-09T14:03:10.7264346Z upload-artifact: 2025-09-09T14:03:10.7264909Z upload-artifact-to-s3: false 2025-09-09T14:03:10.7265188Z download-artifact: 2025-09-09T14:03:10.7265423Z repository: 2025-09-09T14:03:10.7265664Z fetch-depth: 1 2025-09-09T14:03:10.7265892Z submodules: recursive 2025-09-09T14:03:10.7266115Z ref: 2025-09-09T14:03:10.7266359Z test-infra-repository: pytorch/test-infra 2025-09-09T14:03:10.7266660Z test-infra-ref: 2025-09-09T14:03:10.7266911Z use-custom-docker-registry: true 2025-09-09T14:03:10.7267223Z docker-image: pytorch/almalinux-builder 2025-09-09T14:03:10.7267537Z docker-build-dir: .ci/docker 2025-09-09T14:03:10.7267797Z gpu-arch-type: cuda 2025-09-09T14:03:10.7268047Z gpu-arch-version: 12.6 2025-09-09T14:03:10.7268290Z job-name: linux-job 2025-09-09T14:03:10.7268544Z continue-on-error: false 2025-09-09T14:03:10.7268811Z binary-matrix: 2025-09-09T14:03:10.7269031Z run-with-docker: true 2025-09-09T14:03:10.7269276Z secrets-env: 2025-09-09T14:03:10.7269510Z no-sudo: false 2025-09-09T14:03:10.7269740Z ##[endgroup] 2025-09-09T14:03:10.7270284Z Complete job name: test-nightly (CUDA Nightly, linux.g5.12xlarge.nvidia.gpu, --pre torch --index-url https://downloa... / linux-job 2025-09-09T14:03:10.8100643Z A job started hook has been configured by the self-hosted runner administrator 2025-09-09T14:03:10.8237641Z ##[group]Run '/home/ec2-user/runner-scripts/before_job.sh' 2025-09-09T14:03:10.8249152Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:10.8249730Z ##[endgroup] 2025-09-09T14:03:12.3446318Z Runner Type: linux.g5.12xlarge.nvidia.gpu 2025-09-09T14:03:12.3446823Z Instance Type: g5.12xlarge 2025-09-09T14:03:12.3447062Z AMI Name: unknown 2025-09-09T14:03:12.3497260Z AMI ID: ami-05ffe3c48a9991133 2025-09-09T14:03:17.9718743Z ##[group]Run set -euxo pipefail 2025-09-09T14:03:17.9719318Z set -euxo pipefail 2025-09-09T14:03:17.9719631Z if [[ "${NO_SUDO}" == "false" ]]; then 2025-09-09T14:03:17.9720015Z  echo "::group::Cleanup with-sudo debug output" 2025-09-09T14:03:17.9720372Z  sudo rm -rfv "${GITHUB_WORKSPACE}" 2025-09-09T14:03:17.9720653Z else 2025-09-09T14:03:17.9720903Z  echo "::group::Cleanup no-sudo debug output" 2025-09-09T14:03:17.9721236Z  rm -rfv "${GITHUB_WORKSPACE}" 2025-09-09T14:03:17.9721524Z fi 2025-09-09T14:03:17.9721718Z  2025-09-09T14:03:17.9721928Z mkdir -p "${GITHUB_WORKSPACE}" 2025-09-09T14:03:17.9722234Z echo "::endgroup::" 2025-09-09T14:03:17.9737208Z shell: /usr/bin/bash -e {0} 2025-09-09T14:03:17.9737470Z env: 2025-09-09T14:03:17.9737712Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:17.9738043Z REPOSITORY: pytorch/ao 2025-09-09T14:03:17.9738303Z PR_NUMBER: 2963 2025-09-09T14:03:17.9739632Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:17.9741013Z NO_SUDO: false 2025-09-09T14:03:17.9741222Z ##[endgroup] 2025-09-09T14:03:17.9778220Z + [[ false == \f\a\l\s\e ]] 2025-09-09T14:03:17.9790341Z ##[group]Cleanup with-sudo debug output 2025-09-09T14:03:17.9793509Z + echo '::group::Cleanup with-sudo debug output' 2025-09-09T14:03:17.9793918Z + sudo rm -rfv /home/ec2-user/actions-runner/_work/ao/ao 2025-09-09T14:03:18.0931572Z removed directory '/home/ec2-user/actions-runner/_work/ao/ao' 2025-09-09T14:03:18.0954294Z + mkdir -p /home/ec2-user/actions-runner/_work/ao/ao 2025-09-09T14:03:18.0974053Z + echo ::endgroup:: 2025-09-09T14:03:18.0974604Z ##[endgroup] 2025-09-09T14:03:18.1099214Z ##[group]Run actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 2025-09-09T14:03:18.1099641Z with: 2025-09-09T14:03:18.1099858Z repository: pytorch/test-infra 2025-09-09T14:03:18.1100140Z path: test-infra 2025-09-09T14:03:18.1100366Z submodules: recursive 2025-09-09T14:03:18.1100775Z token: *** 2025-09-09T14:03:18.1101003Z ssh-strict: true 2025-09-09T14:03:18.1101214Z ssh-user: git 2025-09-09T14:03:18.1101440Z persist-credentials: true 2025-09-09T14:03:18.1101682Z clean: true 2025-09-09T14:03:18.1101905Z sparse-checkout-cone-mode: true 2025-09-09T14:03:18.1102177Z fetch-depth: 1 2025-09-09T14:03:18.1102382Z fetch-tags: false 2025-09-09T14:03:18.1102600Z show-progress: true 2025-09-09T14:03:18.1102811Z lfs: false 2025-09-09T14:03:18.1103022Z set-safe-directory: true 2025-09-09T14:03:18.1103294Z env: 2025-09-09T14:03:18.1103529Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:18.1103846Z REPOSITORY: pytorch/ao 2025-09-09T14:03:18.1104109Z PR_NUMBER: 2963 2025-09-09T14:03:18.1105430Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:18.1106792Z ##[endgroup] 2025-09-09T14:03:18.2530732Z Syncing repository: pytorch/test-infra 2025-09-09T14:03:18.2531367Z ##[group]Getting Git version info 2025-09-09T14:03:18.2531795Z Working directory is '/home/ec2-user/actions-runner/_work/ao/ao/test-infra' 2025-09-09T14:03:18.2532393Z [command]/usr/bin/git version 2025-09-09T14:03:18.2545391Z git version 2.47.1 2025-09-09T14:03:18.2592816Z ##[endgroup] 2025-09-09T14:03:18.2598205Z Temporarily overriding HOME='/home/ec2-user/actions-runner/_work/_temp/96d2cfcd-5bfe-4ef5-8be8-c619cf6d6d84' before making global git config changes 2025-09-09T14:03:18.2599119Z Adding repository directory to the temporary git global config as a safe directory 2025-09-09T14:03:18.2604538Z [command]/usr/bin/git config --global --add safe.directory /home/ec2-user/actions-runner/_work/ao/ao/test-infra 2025-09-09T14:03:18.2645796Z ##[group]Initializing the repository 2025-09-09T14:03:18.2649957Z [command]/usr/bin/git init /home/ec2-user/actions-runner/_work/ao/ao/test-infra 2025-09-09T14:03:18.2744628Z hint: Using 'master' as the name for the initial branch. This default branch name 2025-09-09T14:03:18.2745211Z hint: is subject to change. To configure the initial branch name to use in all 2025-09-09T14:03:18.2745751Z hint: of your new repositories, which will suppress this warning, call: 2025-09-09T14:03:18.2746151Z hint: 2025-09-09T14:03:18.2746448Z hint: git config --global init.defaultBranch 2025-09-09T14:03:18.2746800Z hint: 2025-09-09T14:03:18.2747136Z hint: Names commonly chosen instead of 'master' are 'main', 'trunk' and 2025-09-09T14:03:18.2747706Z hint: 'development'. The just-created branch can be renamed via this command: 2025-09-09T14:03:18.2748137Z hint: 2025-09-09T14:03:18.2755563Z hint: git branch -m 2025-09-09T14:03:18.2756158Z Initialized empty Git repository in /home/ec2-user/actions-runner/_work/ao/ao/test-infra/.git/ 2025-09-09T14:03:18.2768131Z [command]/usr/bin/git remote add origin https://github.com/pytorch/test-infra 2025-09-09T14:03:18.2860411Z ##[endgroup] 2025-09-09T14:03:18.2861240Z ##[group]Disabling automatic garbage collection 2025-09-09T14:03:18.2864486Z [command]/usr/bin/git config --local gc.auto 0 2025-09-09T14:03:18.2906349Z ##[endgroup] 2025-09-09T14:03:18.2907030Z ##[group]Setting up auth 2025-09-09T14:03:18.2911295Z [command]/usr/bin/git config --local --name-only --get-regexp core\.sshCommand 2025-09-09T14:03:18.2948423Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'core\.sshCommand' && git config --local --unset-all 'core.sshCommand' || :" 2025-09-09T14:03:18.3392138Z [command]/usr/bin/git config --local --name-only --get-regexp http\.https\:\/\/github\.com\/\.extraheader 2025-09-09T14:03:18.3426464Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'http\.https\:\/\/github\.com\/\.extraheader' && git config --local --unset-all 'http.https://github.com/.extraheader' || :" 2025-09-09T14:03:18.3853797Z [command]/usr/bin/git config --local http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:03:18.3906900Z ##[endgroup] 2025-09-09T14:03:18.3907333Z ##[group]Determining the default branch 2025-09-09T14:03:18.3910110Z Retrieving the default branch name 2025-09-09T14:03:18.6723628Z Default branch 'main' 2025-09-09T14:03:18.6724689Z ##[endgroup] 2025-09-09T14:03:18.6725430Z ##[group]Fetching the repository 2025-09-09T14:03:18.6728541Z [command]/usr/bin/git -c protocol.version=2 fetch --no-tags --prune --no-recurse-submodules --depth=1 origin +refs/heads/main:refs/remotes/origin/main 2025-09-09T14:03:19.0750358Z From https://github.com/pytorch/test-infra 2025-09-09T14:03:19.0750767Z * [new branch] main -> origin/main 2025-09-09T14:03:19.0786070Z ##[endgroup] 2025-09-09T14:03:19.0786459Z ##[group]Determining the checkout info 2025-09-09T14:03:19.0787511Z ##[endgroup] 2025-09-09T14:03:19.0792156Z [command]/usr/bin/git sparse-checkout disable 2025-09-09T14:03:19.0843631Z [command]/usr/bin/git config --local --unset-all extensions.worktreeConfig 2025-09-09T14:03:19.0878585Z ##[group]Checking out the ref 2025-09-09T14:03:19.0881655Z [command]/usr/bin/git checkout --progress --force -B main refs/remotes/origin/main 2025-09-09T14:03:19.2455340Z Switched to a new branch 'main' 2025-09-09T14:03:19.2467370Z branch 'main' set up to track 'origin/main'. 2025-09-09T14:03:19.2479660Z ##[endgroup] 2025-09-09T14:03:19.2480061Z ##[group]Setting up auth for fetching submodules 2025-09-09T14:03:19.2485844Z [command]/usr/bin/git config --global http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:03:19.2538849Z [command]/usr/bin/git config --global --unset-all url.https://github.com/.insteadOf 2025-09-09T14:03:19.2582886Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf git@github.com: 2025-09-09T14:03:19.2620298Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf org-21003710@github.com: 2025-09-09T14:03:19.2653663Z ##[endgroup] 2025-09-09T14:03:19.2654055Z ##[group]Fetching submodules 2025-09-09T14:03:19.2657456Z [command]/usr/bin/git submodule sync --recursive 2025-09-09T14:03:19.3064460Z [command]/usr/bin/git -c protocol.version=2 submodule update --init --force --depth=1 --recursive 2025-09-09T14:03:19.3475176Z [command]/usr/bin/git submodule foreach --recursive git config --local gc.auto 0 2025-09-09T14:03:19.3879920Z ##[endgroup] 2025-09-09T14:03:19.3880342Z ##[group]Persisting credentials for submodules 2025-09-09T14:03:19.3884013Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'url\.https\:\/\/github\.com\/\.insteadOf' && git config --local --unset-all 'url.https://github.com/.insteadOf' || :" 2025-09-09T14:03:19.4284364Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local 'http.https://github.com/.extraheader' 'AUTHORIZATION: basic ***' && git config --local --show-origin --name-only --get-regexp remote.origin.url" 2025-09-09T14:03:19.4690027Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'git@github.com:' 2025-09-09T14:03:19.5097388Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'org-21003710@github.com:' 2025-09-09T14:03:19.5498877Z ##[endgroup] 2025-09-09T14:03:19.5542742Z [command]/usr/bin/git log -1 --format=%H 2025-09-09T14:03:19.5575048Z e502b6d9079a2a411c68046e8a7694b851c5df33 2025-09-09T14:03:19.5800004Z Prepare all required actions 2025-09-09T14:03:19.5800426Z Getting action download info 2025-09-09T14:03:19.7357128Z Download action repository 'pytorch/test-infra@main' (SHA:e502b6d9079a2a411c68046e8a7694b851c5df33) 2025-09-09T14:03:21.8513621Z Getting action download info 2025-09-09T14:03:21.9833241Z Download action repository 'nick-fields/retry@3e91a01664abd3c5cd539100d10d33b9c5b68482' (SHA:3e91a01664abd3c5cd539100d10d33b9c5b68482) 2025-09-09T14:03:22.1566558Z ##[group]Run ./test-infra/.github/actions/setup-linux 2025-09-09T14:03:22.1566884Z env: 2025-09-09T14:03:22.1567125Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.1567445Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.1567679Z PR_NUMBER: 2963 2025-09-09T14:03:22.1568997Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.1570314Z ##[endgroup] 2025-09-09T14:03:22.1648608Z ##[group]Run set -euo pipefail 2025-09-09T14:03:22.1648912Z set -euo pipefail 2025-09-09T14:03:22.1649178Z function get_ec2_metadata() { 2025-09-09T14:03:22.1649516Z  # Pulled from instance metadata endpoint for EC2 2025-09-09T14:03:22.1650098Z  # see https://docs.aws.amazon.com/AWSEC2/latest/UserGuide/instancedata-data-retrieval.html 2025-09-09T14:03:22.1650604Z  category=$1 2025-09-09T14:03:22.1651418Z  curl -H "X-aws-ec2-metadata-token: $(curl -s -X PUT "http://169.254.169.254/latest/api/token" -H "X-aws-ec2-metadata-token-ttl-seconds: 30")" -fsSL "http://169.254.169.254/latest/meta-data/${category}" 2025-09-09T14:03:22.1652233Z } 2025-09-09T14:03:22.1652472Z echo "ami-id: $(get_ec2_metadata ami-id)" 2025-09-09T14:03:22.1652872Z echo "instance-id: $(get_ec2_metadata instance-id)" 2025-09-09T14:03:22.1653308Z echo "instance-type: $(get_ec2_metadata instance-type)" 2025-09-09T14:03:22.1654957Z echo "system info $(uname -a)" 2025-09-09T14:03:22.1664548Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:22.1664893Z env: 2025-09-09T14:03:22.1665133Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.1665466Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.1665715Z PR_NUMBER: 2963 2025-09-09T14:03:22.1667011Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.1668354Z ##[endgroup] 2025-09-09T14:03:22.1830975Z ami-id: ami-05ffe3c48a9991133 2025-09-09T14:03:22.1955641Z instance-id: i-0164110f67924097b 2025-09-09T14:03:22.2084948Z instance-type: g5.12xlarge 2025-09-09T14:03:22.2116877Z system info Linux ip-10-0-57-110.ec2.internal 6.1.141-155.222.amzn2023.x86_64 #1 SMP PREEMPT_DYNAMIC Tue Jun 17 10:29:47 UTC 2025 x86_64 x86_64 x86_64 GNU/Linux 2025-09-09T14:03:22.2148623Z ##[group]Run echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:03:22.2149442Z echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:03:22.2157949Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:22.2158305Z env: 2025-09-09T14:03:22.2158554Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.2158877Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.2159130Z PR_NUMBER: 2963 2025-09-09T14:03:22.2160635Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.2161986Z ##[endgroup] 2025-09-09T14:03:22.2244642Z ##[group]Run if systemctl is-active --quiet docker; then 2025-09-09T14:03:22.2245045Z if systemctl is-active --quiet docker; then 2025-09-09T14:03:22.2245390Z  echo "Docker daemon is running..."; 2025-09-09T14:03:22.2245692Z else 2025-09-09T14:03:22.2246008Z  echo "Starting docker deamon..." && sudo systemctl start docker; 2025-09-09T14:03:22.2246396Z fi 2025-09-09T14:03:22.2255595Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:22.2255943Z env: 2025-09-09T14:03:22.2256207Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.2256555Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.2256814Z PR_NUMBER: 2963 2025-09-09T14:03:22.2258116Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.2259486Z ##[endgroup] 2025-09-09T14:03:22.2357676Z Docker daemon is running... 2025-09-09T14:03:22.2390824Z ##[group]Run AWS_ACCOUNT_ID=$(aws sts get-caller-identity|grep Account|cut -f4 -d\") 2025-09-09T14:03:22.2391395Z AWS_ACCOUNT_ID=$(aws sts get-caller-identity|grep Account|cut -f4 -d\") 2025-09-09T14:03:22.2391859Z retry () { "$@" || (sleep 1 && "$@") || (sleep 2 && "$@") } 2025-09-09T14:03:22.2392410Z retry aws ecr get-login-password --region "$AWS_DEFAULT_REGION" | docker login --username AWS \ 2025-09-09T14:03:22.2393063Z  --password-stdin "$AWS_ACCOUNT_ID.dkr.ecr.$AWS_DEFAULT_REGION.amazonaws.com" 2025-09-09T14:03:22.2401683Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:22.2402015Z env: 2025-09-09T14:03:22.2402263Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:22.2402587Z REPOSITORY: pytorch/ao 2025-09-09T14:03:22.2402831Z PR_NUMBER: 2963 2025-09-09T14:03:22.2404115Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:22.2405451Z AWS_RETRY_MODE: standard 2025-09-09T14:03:22.2405698Z AWS_MAX_ATTEMPTS: 5 2025-09-09T14:03:22.2405928Z AWS_DEFAULT_REGION: us-east-1 2025-09-09T14:03:22.2406185Z ##[endgroup] 2025-09-09T14:03:23.3010682Z WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json. 2025-09-09T14:03:23.3011455Z Configure a credential helper to remove this warning. See 2025-09-09T14:03:23.3012140Z https://docs.docker.com/engine/reference/commandline/login/#credentials-store 2025-09-09T14:03:23.3012609Z 2025-09-09T14:03:23.3012794Z Login Succeeded 2025-09-09T14:03:23.3071831Z ##[group]Run env | grep '^GITHUB' >> "${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID}" 2025-09-09T14:03:23.3072357Z env | grep '^GITHUB' >> "${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID}" 2025-09-09T14:03:23.3072809Z env | grep '^CI' >> "${RUNNER_TEMP}/github_env_${GITHUB_RUN_ID}" 2025-09-09T14:03:23.3082835Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:23.3083208Z env: 2025-09-09T14:03:23.3083470Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.3083831Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.3084332Z PR_NUMBER: 2963 2025-09-09T14:03:23.3085634Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.3086977Z ##[endgroup] 2025-09-09T14:03:23.3207669Z ##[group]Run RUNNER_ARTIFACT_DIR="${RUNNER_TEMP}/artifacts" 2025-09-09T14:03:23.3208114Z RUNNER_ARTIFACT_DIR="${RUNNER_TEMP}/artifacts" 2025-09-09T14:03:23.3208467Z sudo rm -rf "${RUNNER_ARTIFACT_DIR}" 2025-09-09T14:03:23.3208790Z mkdir -p "${RUNNER_ARTIFACT_DIR}" 2025-09-09T14:03:23.3209185Z echo "RUNNER_ARTIFACT_DIR=${RUNNER_ARTIFACT_DIR}" >> "${GITHUB_ENV}" 2025-09-09T14:03:23.3209564Z  2025-09-09T14:03:23.3209854Z RUNNER_TEST_RESULTS_DIR="${RUNNER_TEMP}/test-results" 2025-09-09T14:03:23.3210237Z sudo rm -rf "${RUNNER_TEST_RESULTS_DIR}" 2025-09-09T14:03:23.3210570Z mkdir -p "${RUNNER_TEST_RESULTS_DIR}" 2025-09-09T14:03:23.3211014Z echo "RUNNER_TEST_RESULTS_DIR=${RUNNER_TEST_RESULTS_DIR}" >> "${GITHUB_ENV}" 2025-09-09T14:03:23.3211417Z  2025-09-09T14:03:23.3211631Z RUNNER_DOCS_DIR="${RUNNER_TEMP}/docs" 2025-09-09T14:03:23.3211942Z sudo rm -rf "${RUNNER_DOCS_DIR}" 2025-09-09T14:03:23.3212237Z mkdir -p "${RUNNER_DOCS_DIR}" 2025-09-09T14:03:23.3212595Z echo "RUNNER_DOCS_DIR=${RUNNER_DOCS_DIR}" >> "${GITHUB_ENV}" 2025-09-09T14:03:23.3222058Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:23.3222379Z env: 2025-09-09T14:03:23.3222617Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.3222936Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.3223176Z PR_NUMBER: 2963 2025-09-09T14:03:23.3224468Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.3226020Z ##[endgroup] 2025-09-09T14:03:23.8819173Z ##[group]Run needs=0 2025-09-09T14:03:23.8819419Z needs=0 2025-09-09T14:03:23.8819753Z if lspci -v | grep -e 'controller.*NVIDIA' >/dev/null 2>/dev/null; then 2025-09-09T14:03:23.8820153Z  needs=1 2025-09-09T14:03:23.8820366Z fi 2025-09-09T14:03:23.8820595Z echo "does=${needs}" >> $GITHUB_OUTPUT 2025-09-09T14:03:23.8830327Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:03:23.8830663Z env: 2025-09-09T14:03:23.8830955Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.8831274Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.8831527Z PR_NUMBER: 2963 2025-09-09T14:03:23.8832819Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.8834288Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:03:23.8834821Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:03:23.8835314Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:03:23.8835667Z ##[endgroup] 2025-09-09T14:03:23.9180832Z ##[group]Run pytorch/test-infra/.github/actions/setup-nvidia@main 2025-09-09T14:03:23.9181192Z with: 2025-09-09T14:03:23.9181388Z driver-version: 580.65.06 2025-09-09T14:03:23.9181621Z env: 2025-09-09T14:03:23.9182053Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.9182373Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.9182604Z PR_NUMBER: 2963 2025-09-09T14:03:23.9183890Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.9185348Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:03:23.9185885Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:03:23.9186371Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:03:23.9186733Z ##[endgroup] 2025-09-09T14:03:23.9230159Z ##[group]Run nick-fields/retry@3e91a01664abd3c5cd539100d10d33b9c5b68482 2025-09-09T14:03:23.9230532Z with: 2025-09-09T14:03:23.9230737Z timeout_minutes: 10 2025-09-09T14:03:23.9230957Z max_attempts: 3 2025-09-09T14:03:23.9255316Z command: # Is it disgusting to have a full shell script here in this github action? Sure # But is it the best way to make it so that this action relies on nothing else? Absolutely set -eou pipefail DISTRIBUTION=$(. /etc/os-release;echo $ID$VERSION_ID) DRIVER_FN="NVIDIA-Linux-x86_64-${DRIVER_VERSION}.run" install_nvidia_docker2_amzn2() { ( set -x # Needed for yum-config-manager sudo yum install -y yum-utils if [[ "${DISTRIBUTION}" == "amzn2023" ]] ; then YUM_REPO_URL="https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo" else # Amazon Linux 2 YUM_REPO_URL="https://nvidia.github.io/nvidia-docker/${DISTRIBUTION}/nvidia-docker.repo" fi sudo yum-config-manager --add-repo "${YUM_REPO_URL}" sudo yum install -y \ nvidia-container-toolkit-1.17.8 \ libnvidia-container-tools-1.17.8 \ libnvidia-container1-1.17.8 \ nvidia-container-toolkit-base-1.17.8 sudo systemctl restart docker ) } install_nvidia_docker2_ubuntu20() { ( set -x # Install nvidia-driver package if not installed status="$(dpkg-query -W --showformat='${db:Status-Status}' nvidia-docker2 2>&1)" if [ ! $? = 0 ] || [ ! "$status" = installed ]; then sudo apt-get install -y nvidia-container-toolkit-1.17.8 sudo systemctl restart docker fi ) } pre_install_nvidia_driver_amzn2() { ( # Purge any nvidia driver installed from RHEL repo sudo yum remove -y nvidia-driver-latest-dkms ) } install_nvidia_driver_common() { ( # Try to gather more information about the runner and its existing NVIDIA driver if any echo "Before installing NVIDIA driver" lspci lsmod modinfo nvidia || true HAS_NVIDIA_DRIVER=0 # Check if NVIDIA driver has already been installed if [ -x "$(command -v nvidia-smi)" ]; then set +e # The driver exists, check its version next. Also check only the first GPU if there are more than one of them # so that the same driver version is not print over multiple lines INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then echo "Failed to get NVIDIA driver version ($INSTALLED_DRIVER_VERSION). Continuing" elif [ "$INSTALLED_DRIVER_VERSION" != "$DRIVER_VERSION" ]; then echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has been installed, but we expect to have $DRIVER_VERSION instead. Continuing" # Turn off persistent mode so that the installation script can unload the kernel module sudo killall nvidia-persistenced || true else HAS_NVIDIA_DRIVER=1 echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has already been installed. Skipping NVIDIA driver installation" fi set -e fi if [ "$HAS_NVIDIA_DRIVER" -eq 0 ]; then # CAUTION: this may need to be updated in future if [ "${DISTRIBUTION}" != ubuntu20.04 ]; then sudo yum groupinstall -y "Development Tools" # ensure our kernel install is the same as our underlying kernel, # groupinstall "Development Tools" has a habit of mismatching kernel headers sudo yum install -y "kernel-devel-uname-r == $(uname -r)" sudo modprobe backlight fi sudo curl -fsL -o /tmp/nvidia_driver "https://s3.amazonaws.com/ossci-linux/nvidia_driver/$DRIVER_FN" set +e sudo /bin/bash /tmp/nvidia_driver -s --no-drm NVIDIA_INSTALLATION_STATUS=$? RESET_GPU=0 if [ "$NVIDIA_INSTALLATION_STATUS" -ne 0 ]; then sudo cat /var/log/nvidia-installer.log # Fail to install NVIDIA driver, try to reset the GPU RESET_GPU=1 elif [ -x "$(command -v nvidia-smi)" ]; then # Check again if nvidia-smi works even if the driver installation completes successfully INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then RESET_GPU=1 fi fi if [ "$RESET_GPU" -eq 1 ]; then NVIDIA_DEVICES=$(lspci -D | grep -i NVIDIA | cut -d' ' -f1) # The GPU can get stuck in a failure state if somehow the test crashs the GPU microcode. When this # happens, we'll try to reset all NVIDIA devices https://github.com/pytorch/pytorch/issues/88388 for PCI_ID in $NVIDIA_DEVICES; do DEVICE_ENABLED=$(cat /sys/bus/pci/devices/$PCI_ID/enable) echo "Reseting $PCI_ID (enabled state: $DEVICE_ENABLED)" # This requires sudo permission of course echo "1" | sudo tee /sys/bus/pci/devices/$PCI_ID/reset sleep 1 done fi sudo rm -fv /tmp/nvidia_driver set -e fi ) } post_install_nvidia_driver_common() { ( sudo modprobe nvidia || true echo "After installing NVIDIA driver" lspci lsmod modinfo nvidia || true ( set +e nvidia-smi # NB: Annoyingly, nvidia-smi command returns successfully with return code 0 even in # the case where the driver has already crashed as it still can get the driver version # and some basic information like the bus ID. However, the rest of the information # would be missing (ERR!), for example: # # +-----------------------------------------------------------------------------+ # | NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 | # |-------------------------------+----------------------+----------------------+ # | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | # | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | # | | | MIG M. | # |===============================+======================+======================| # | 0 ERR! Off | 00000000:00:1E.0 Off | ERR! | # |ERR! ERR! ERR! ERR! / ERR! | 4184MiB / 23028MiB | ERR! Default | # | | | ERR! | # +-------------------------------+----------------------+----------------------+ # # +-----------------------------------------------------------------------------+ # | Processes: | # | GPU GI CI PID Type Process name GPU Memory | # | ID ID Usage | # |=============================================================================| # +-----------------------------------------------------------------------------+ # # This should be reported as a failure instead as it will guarantee to fail when # Docker tries to run with --gpus all # # So, the correct check here is to query one of the missing piece of info like # GPU name, so that the command can fail accordingly nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 NVIDIA_SMI_STATUS=$? # Allowable exit statuses for nvidia-smi, see: https://github.com/NVIDIA/gpu-operator/issues/285 if [ "$NVIDIA_SMI_STATUS" -eq 0 ] || [ "$NVIDIA_SMI_STATUS" -eq 14 ]; then echo "INFO: Ignoring allowed status ${NVIDIA_SMI_STATUS}" else echo "ERROR: nvidia-smi exited with unresolved status ${NVIDIA_SMI_STATUS}" exit ${NVIDIA_SMI_STATUS} fi set -e ) ) } install_nvidia_driver_amzn2() { ( set -x pre_install_nvidia_driver_amzn2 install_nvidia_driver_common post_install_nvidia_driver_common ) } install_nvidia_driver_ubuntu20() { ( set -x install_nvidia_driver_common post_install_nvidia_driver_common ) } echo "== Installing nvidia driver ${DRIVER_FN} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_driver_amzn2 ;; ubuntu20.04) install_nvidia_driver_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac # Install container toolkit based on distribution echo "== Installing nvidia container toolkit for ${DISTRIBUTION} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_docker2_amzn2 ;; ubuntu20.04) install_nvidia_docker2_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac echo "GPU_FLAG=--gpus all -e NVIDIA_DRIVER_CAPABILITIES=all" >> "${GITHUB_ENV}" # Fix https://github.com/NVIDIA/nvidia-docker/issues/1648 on runners with # more than one GPUs. This just needs to be run once. The command fails # on subsequent runs and complains that the mode is already on, but that's # ok sudo nvidia-persistenced || true # This should show persistence mode ON nvidia-smi # check if the container-toolkit is correctly installed and CUDA is available inside a container docker run --rm -t --gpus=all public.ecr.aws/docker/library/python:3.13 nvidia-smi 2025-09-09T14:03:23.9280471Z retry_wait_seconds: 10 2025-09-09T14:03:23.9280718Z polling_interval_seconds: 1 2025-09-09T14:03:23.9280993Z warning_on_retry: true 2025-09-09T14:03:23.9281229Z continue_on_error: false 2025-09-09T14:03:23.9281467Z env: 2025-09-09T14:03:23.9281700Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:03:23.9282035Z REPOSITORY: pytorch/ao 2025-09-09T14:03:23.9282276Z PR_NUMBER: 2963 2025-09-09T14:03:23.9283594Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:03:23.9285089Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:03:23.9285775Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:03:23.9286281Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:03:23.9286653Z DRIVER_VERSION: 580.65.06 2025-09-09T14:03:23.9286890Z ##[endgroup] 2025-09-09T14:03:24.0112082Z == Installing nvidia driver NVIDIA-Linux-x86_64-580.65.06.run == 2025-09-09T14:03:24.0112894Z + pre_install_nvidia_driver_amzn2 2025-09-09T14:03:24.0118245Z + sudo yum remove -y nvidia-driver-latest-dkms 2025-09-09T14:03:24.3894273Z No match for argument: nvidia-driver-latest-dkms 2025-09-09T14:03:24.3895417Z No packages marked for removal. 2025-09-09T14:03:24.3959210Z Dependencies resolved. 2025-09-09T14:03:24.3969705Z Nothing to do. 2025-09-09T14:03:24.3970126Z Complete! 2025-09-09T14:03:24.4304902Z + install_nvidia_driver_common 2025-09-09T14:03:24.4309329Z + echo 'Before installing NVIDIA driver' 2025-09-09T14:03:24.4309616Z + lspci 2025-09-09T14:03:24.4312536Z Before installing NVIDIA driver 2025-09-09T14:03:24.4444753Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:03:24.4445252Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:03:24.4445814Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:03:24.4446323Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:03:24.4446784Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:03:24.4447294Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:03:24.4447762Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:24.4448175Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:24.4448588Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:24.4448991Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:03:24.4449453Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:03:24.4449833Z + lsmod 2025-09-09T14:03:24.4498501Z Module Size Used by 2025-09-09T14:03:24.4499144Z xt_conntrack 16384 1 2025-09-09T14:03:24.4499394Z nft_chain_nat 16384 3 2025-09-09T14:03:24.4499653Z xt_MASQUERADE 20480 1 2025-09-09T14:03:24.4499939Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:03:24.4500266Z nf_conntrack_netlink 57344 0 2025-09-09T14:03:24.4500648Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:03:24.4501075Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:03:24.4501383Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:03:24.4501663Z xfrm_user 57344 1 2025-09-09T14:03:24.4501929Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:03:24.4502205Z xt_addrtype 16384 2 2025-09-09T14:03:24.4502462Z nft_compat 20480 4 2025-09-09T14:03:24.4502759Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:03:24.4503165Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:03:24.4503533Z br_netfilter 36864 0 2025-09-09T14:03:24.4503808Z bridge 323584 1 br_netfilter 2025-09-09T14:03:24.4504097Z stp 16384 1 bridge 2025-09-09T14:03:24.4504371Z llc 16384 2 bridge,stp 2025-09-09T14:03:24.4504650Z overlay 167936 0 2025-09-09T14:03:24.4504888Z tls 139264 0 2025-09-09T14:03:24.4505130Z nls_ascii 16384 1 2025-09-09T14:03:24.4505371Z nls_cp437 20480 1 2025-09-09T14:03:24.4505616Z vfat 24576 1 2025-09-09T14:03:24.4505853Z fat 86016 1 vfat 2025-09-09T14:03:24.4506111Z sunrpc 700416 1 2025-09-09T14:03:24.4506351Z i8042 45056 0 2025-09-09T14:03:24.4506582Z ena 180224 0 2025-09-09T14:03:24.4506979Z serio 28672 3 i8042 2025-09-09T14:03:24.4507243Z ghash_clmulni_intel 16384 0 2025-09-09T14:03:24.4507497Z button 24576 0 2025-09-09T14:03:24.4507742Z sch_fq_codel 20480 33 2025-09-09T14:03:24.4507995Z fuse 184320 1 2025-09-09T14:03:24.4508227Z dm_mod 188416 0 2025-09-09T14:03:24.4508469Z loop 36864 0 2025-09-09T14:03:24.4508702Z configfs 57344 1 2025-09-09T14:03:24.4508945Z dmi_sysfs 20480 0 2025-09-09T14:03:24.4509189Z crc32_pclmul 16384 0 2025-09-09T14:03:24.4509434Z crc32c_intel 24576 0 2025-09-09T14:03:24.4509680Z efivarfs 24576 1 2025-09-09T14:03:24.4509920Z + modinfo nvidia 2025-09-09T14:03:24.4526484Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:03:24.4526944Z import_ns: DMA_BUF 2025-09-09T14:03:24.4527181Z alias: char-major-195-* 2025-09-09T14:03:24.4527450Z version: 570.133.07 2025-09-09T14:03:24.4527685Z supported: external 2025-09-09T14:03:24.4527928Z license: Dual MIT/GPL 2025-09-09T14:03:24.4528207Z firmware: nvidia/570.133.07/gsp_tu10x.bin 2025-09-09T14:03:24.4528539Z firmware: nvidia/570.133.07/gsp_ga10x.bin 2025-09-09T14:03:24.4528870Z srcversion: 49515739FD8F721A3F2F714 2025-09-09T14:03:24.4529179Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:03:24.4529507Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:03:24.4529826Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:03:24.4530131Z depends: i2c-core,drm 2025-09-09T14:03:24.4530374Z retpoline: Y 2025-09-09T14:03:24.4530587Z name: nvidia 2025-09-09T14:03:24.4530929Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:03:24.4531396Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:03:24.4531859Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:03:24.4532267Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:03:24.4532572Z parm: NVreg_RmLogonRC:int 2025-09-09T14:03:24.4532976Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:03:24.4533290Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:03:24.4533579Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:03:24.4533881Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:03:24.4534235Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:03:24.4534615Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:03:24.4535003Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:03:24.4535527Z parm: NVreg_EnableMSI:int 2025-09-09T14:03:24.4535916Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:03:24.4536362Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:03:24.4550223Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:03:24.4550627Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:03:24.4551039Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:03:24.4551459Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:03:24.4551875Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:03:24.4552289Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:03:24.4552618Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:03:24.4552987Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:03:24.4553361Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:03:24.4553691Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:03:24.4554016Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:03:24.4554338Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:03:24.4554666Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:03:24.4554970Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:03:24.4555448Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:03:24.4555805Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:03:24.4556289Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:03:24.4556756Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:03:24.4557213Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:03:24.4557665Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:03:24.4558115Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:03:24.4558554Z parm: NVreg_RmMsg:charp 2025-09-09T14:03:24.4558923Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:03:24.4559272Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:03:24.4559588Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:03:24.4559899Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:03:24.4560222Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:03:24.4560579Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:03:24.4560935Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:03:24.4561253Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:03:24.4561603Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:03:24.4561937Z parm: rm_firmware_active:charp 2025-09-09T14:03:24.4562237Z + HAS_NVIDIA_DRIVER=0 2025-09-09T14:03:24.4562475Z ++ command -v nvidia-smi 2025-09-09T14:03:24.4562737Z + '[' -x /usr/bin/nvidia-smi ']' 2025-09-09T14:03:24.4562993Z + set +e 2025-09-09T14:03:24.4563296Z ++ nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0 2025-09-09T14:03:27.8576920Z + INSTALLED_DRIVER_VERSION=570.133.07 2025-09-09T14:03:27.8577374Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:03:27.8577695Z + '[' 0 -ne 0 ']' 2025-09-09T14:03:27.8577946Z + '[' 570.133.07 '!=' 580.65.06 ']' 2025-09-09T14:03:27.8578411Z + echo 'NVIDIA driver (570.133.07) has been installed, but we expect to have 580.65.06 instead. Continuing' 2025-09-09T14:03:27.8578931Z + sudo killall nvidia-persistenced 2025-09-09T14:03:27.8579390Z NVIDIA driver (570.133.07) has been installed, but we expect to have 580.65.06 instead. Continuing 2025-09-09T14:03:27.9450185Z nvidia-persistenced: no process found 2025-09-09T14:03:27.9472193Z + true 2025-09-09T14:03:27.9472468Z + set -e 2025-09-09T14:03:27.9472713Z + '[' 0 -eq 0 ']' 2025-09-09T14:03:27.9473018Z + '[' amzn2023 '!=' ubuntu20.04 ']' 2025-09-09T14:03:27.9473434Z + sudo yum groupinstall -y 'Development Tools' 2025-09-09T14:03:28.4384383Z Last metadata expiration check: 0:06:28 ago on Tue Sep 9 13:57:00 2025. 2025-09-09T14:03:28.4728409Z No match for group package "system-rpm-config" 2025-09-09T14:03:28.4742737Z No match for group package "rcs" 2025-09-09T14:03:28.4759929Z No match for group package "pkgconfig" 2025-09-09T14:03:28.5207785Z Dependencies resolved. 2025-09-09T14:03:28.5426830Z ================================================================================ 2025-09-09T14:03:28.5427412Z Package Architecture Version Repository Size 2025-09-09T14:03:28.5427828Z ================================================================================ 2025-09-09T14:03:28.5428142Z Installing Groups: 2025-09-09T14:03:28.5428442Z Development Tools 2025-09-09T14:03:28.5428710Z 2025-09-09T14:03:28.5428807Z Transaction Summary 2025-09-09T14:03:28.5429041Z ================================================================================ 2025-09-09T14:03:28.5429260Z 2025-09-09T14:03:28.7536890Z ================================================================================ 2025-09-09T14:03:28.7537277Z WARNING: 2025-09-09T14:03:28.7537522Z A newer release of "Amazon Linux" is available. 2025-09-09T14:03:28.7537745Z 2025-09-09T14:03:28.7537838Z Available Versions: 2025-09-09T14:03:28.7537986Z 2025-09-09T14:03:28.7538084Z Version 2023.8.20250707: 2025-09-09T14:03:28.7538389Z Run the following command to upgrade to 2023.8.20250707: 2025-09-09T14:03:28.7538863Z 2025-09-09T14:03:28.7538987Z dnf upgrade --releasever=2023.8.20250707 2025-09-09T14:03:28.7539194Z 2025-09-09T14:03:28.7539279Z Release notes: 2025-09-09T14:03:28.7539700Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250707.html 2025-09-09T14:03:28.7540063Z 2025-09-09T14:03:28.7540162Z Version 2023.8.20250715: 2025-09-09T14:03:28.7540462Z Run the following command to upgrade to 2023.8.20250715: 2025-09-09T14:03:28.7540720Z 2025-09-09T14:03:28.7540837Z dnf upgrade --releasever=2023.8.20250715 2025-09-09T14:03:28.7541042Z 2025-09-09T14:03:28.7541127Z Release notes: 2025-09-09T14:03:28.7541522Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250715.html 2025-09-09T14:03:28.7541880Z 2025-09-09T14:03:28.7541975Z Version 2023.8.20250721: 2025-09-09T14:03:28.7542271Z Run the following command to upgrade to 2023.8.20250721: 2025-09-09T14:03:28.7542512Z 2025-09-09T14:03:28.7542640Z dnf upgrade --releasever=2023.8.20250721 2025-09-09T14:03:28.7542842Z 2025-09-09T14:03:28.7542924Z Release notes: 2025-09-09T14:03:28.7543313Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250721.html 2025-09-09T14:03:28.7543675Z 2025-09-09T14:03:28.7543766Z Version 2023.8.20250808: 2025-09-09T14:03:28.7544066Z Run the following command to upgrade to 2023.8.20250808: 2025-09-09T14:03:28.7544307Z 2025-09-09T14:03:28.7544426Z dnf upgrade --releasever=2023.8.20250808 2025-09-09T14:03:28.7544628Z 2025-09-09T14:03:28.7544711Z Release notes: 2025-09-09T14:03:28.7545097Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250808.html 2025-09-09T14:03:28.7545451Z 2025-09-09T14:03:28.7545566Z Version 2023.8.20250818: 2025-09-09T14:03:28.7545860Z Run the following command to upgrade to 2023.8.20250818: 2025-09-09T14:03:28.7546101Z 2025-09-09T14:03:28.7546220Z dnf upgrade --releasever=2023.8.20250818 2025-09-09T14:03:28.7546422Z 2025-09-09T14:03:28.7546511Z Release notes: 2025-09-09T14:03:28.7546900Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250818.html 2025-09-09T14:03:28.7547526Z 2025-09-09T14:03:28.7547616Z Version 2023.8.20250908: 2025-09-09T14:03:28.7547915Z Run the following command to upgrade to 2023.8.20250908: 2025-09-09T14:03:28.7548156Z 2025-09-09T14:03:28.7548275Z dnf upgrade --releasever=2023.8.20250908 2025-09-09T14:03:28.7548478Z 2025-09-09T14:03:28.7548563Z Release notes: 2025-09-09T14:03:28.7548957Z https://docs.aws.amazon.com/linux/al2023/release-notes/relnotes-2023.8.20250908.html 2025-09-09T14:03:28.7549312Z 2025-09-09T14:03:28.7549420Z ================================================================================ 2025-09-09T14:03:28.7549730Z Complete! 2025-09-09T14:03:28.7974886Z ++ uname -r 2025-09-09T14:03:28.7988514Z + sudo yum install -y 'kernel-devel-uname-r == 6.1.141-155.222.amzn2023.x86_64' 2025-09-09T14:03:29.2738171Z Last metadata expiration check: 0:06:29 ago on Tue Sep 9 13:57:00 2025. 2025-09-09T14:03:29.2964947Z Using '==' operator in reldeps can result in an undefined behavior. It is deprecated and the support will be dropped in future versions. Use '=' operator instead. 2025-09-09T14:03:29.3079881Z Package kernel-devel-6.1.141-155.222.amzn2023.x86_64 is already installed. 2025-09-09T14:03:29.3545780Z Dependencies resolved. 2025-09-09T14:03:29.3770105Z Nothing to do. 2025-09-09T14:03:29.3770767Z Complete! 2025-09-09T14:03:29.4174309Z + sudo modprobe backlight 2025-09-09T14:03:29.5569369Z + sudo curl -fsL -o /tmp/nvidia_driver https://s3.amazonaws.com/ossci-linux/nvidia_driver/NVIDIA-Linux-x86_64-580.65.06.run 2025-09-09T14:03:33.7044182Z + set +e 2025-09-09T14:03:33.7044604Z + sudo /bin/bash /tmp/nvidia_driver -s --no-drm 2025-09-09T14:03:35.0135187Z Verifying archive integrity... OK 2025-09-09T14:03:37.8229030Z Uncompressing NVIDIA Accelerated Graphics Driver for Linux-x86_64 580.65.06.................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................................... 2025-09-09T14:03:38.4549141Z 2025-09-09T14:03:38.4549973Z WARNING: The nvidia-drm module will not be installed. As a result, DRM-KMS will not function with this installation of the NVIDIA driver. 2025-09-09T14:03:38.4550485Z 2025-09-09T14:04:00.9449160Z 2025-09-09T14:04:00.9450701Z WARNING: nvidia-installer was forced to guess the X library path '/usr/lib64' and X module path '/usr/lib64/xorg/modules'; these paths were not queryable from the system. If X fails to find the NVIDIA X driver module, please install the `pkg-config` utility and the X.Org SDK/development package for your distribution and reinstall the driver. 2025-09-09T14:04:00.9451915Z 2025-09-09T14:04:00.9471692Z 2025-09-09T14:04:00.9472927Z WARNING: This NVIDIA driver package includes Vulkan components, but no Vulkan ICD loader was detected on this system. The NVIDIA Vulkan ICD will not function without the loader. Most distributions package the Vulkan loader; try installing the "vulkan-loader", "vulkan-icd-loader", or "libvulkan1" package. 2025-09-09T14:04:00.9474008Z 2025-09-09T14:04:13.1576499Z + NVIDIA_INSTALLATION_STATUS=0 2025-09-09T14:04:13.1576824Z + RESET_GPU=0 2025-09-09T14:04:13.1577037Z + '[' 0 -ne 0 ']' 2025-09-09T14:04:13.1579021Z ++ command -v nvidia-smi 2025-09-09T14:04:13.1582460Z + '[' -x /usr/bin/nvidia-smi ']' 2025-09-09T14:04:13.1588243Z ++ nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0 2025-09-09T14:04:16.9931155Z + INSTALLED_DRIVER_VERSION=580.65.06 2025-09-09T14:04:16.9931493Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:04:16.9931724Z + '[' 0 -ne 0 ']' 2025-09-09T14:04:16.9931932Z + '[' 0 -eq 1 ']' 2025-09-09T14:04:16.9932149Z + sudo rm -fv /tmp/nvidia_driver 2025-09-09T14:04:17.1587080Z removed '/tmp/nvidia_driver' 2025-09-09T14:04:17.1614981Z + set -e 2025-09-09T14:04:17.1619948Z + post_install_nvidia_driver_common 2025-09-09T14:04:17.1624002Z + sudo modprobe nvidia 2025-09-09T14:04:17.4976822Z + echo 'After installing NVIDIA driver' 2025-09-09T14:04:17.4977125Z + lspci 2025-09-09T14:04:17.4977334Z After installing NVIDIA driver 2025-09-09T14:04:17.5109355Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:04:17.5109832Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:04:17.5110360Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:04:17.5110858Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:04:17.5111321Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:04:17.5111833Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:04:17.5112291Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5112706Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5113404Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5113814Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:04:17.5114261Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:04:17.5114655Z + lsmod 2025-09-09T14:04:17.5151550Z Module Size Used by 2025-09-09T14:04:17.5151841Z nvidia_uvm 1921024 0 2025-09-09T14:04:17.5152101Z nvidia 14274560 1 nvidia_uvm 2025-09-09T14:04:17.5152375Z drm 602112 1 nvidia 2025-09-09T14:04:17.5152665Z drm_panel_orientation_quirks 32768 1 drm 2025-09-09T14:04:17.5152959Z backlight 24576 1 drm 2025-09-09T14:04:17.5153230Z i2c_core 110592 2 nvidia,drm 2025-09-09T14:04:17.5153501Z xt_conntrack 16384 1 2025-09-09T14:04:17.5153751Z nft_chain_nat 16384 3 2025-09-09T14:04:17.5154001Z xt_MASQUERADE 20480 1 2025-09-09T14:04:17.5154281Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:04:17.5154609Z nf_conntrack_netlink 57344 0 2025-09-09T14:04:17.5154988Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:04:17.5155411Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:04:17.5155700Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:04:17.5156065Z xfrm_user 57344 1 2025-09-09T14:04:17.5156314Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:04:17.5156591Z xt_addrtype 16384 2 2025-09-09T14:04:17.5156835Z nft_compat 20480 4 2025-09-09T14:04:17.5157123Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:04:17.5157535Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:04:17.5157889Z br_netfilter 36864 0 2025-09-09T14:04:17.5158156Z bridge 323584 1 br_netfilter 2025-09-09T14:04:17.5158430Z stp 16384 1 bridge 2025-09-09T14:04:17.5158704Z llc 16384 2 bridge,stp 2025-09-09T14:04:17.5158972Z overlay 167936 0 2025-09-09T14:04:17.5159202Z tls 139264 0 2025-09-09T14:04:17.5159618Z nls_ascii 16384 1 2025-09-09T14:04:17.5159850Z nls_cp437 20480 1 2025-09-09T14:04:17.5160084Z vfat 24576 1 2025-09-09T14:04:17.5160314Z fat 86016 1 vfat 2025-09-09T14:04:17.5160568Z sunrpc 700416 1 2025-09-09T14:04:17.5160795Z i8042 45056 0 2025-09-09T14:04:17.5161026Z ena 180224 0 2025-09-09T14:04:17.5161268Z serio 28672 3 i8042 2025-09-09T14:04:17.5161523Z ghash_clmulni_intel 16384 0 2025-09-09T14:04:17.5161766Z button 24576 0 2025-09-09T14:04:17.5162000Z sch_fq_codel 20480 33 2025-09-09T14:04:17.5162241Z fuse 184320 1 2025-09-09T14:04:17.5162470Z dm_mod 188416 0 2025-09-09T14:04:17.5162701Z loop 36864 0 2025-09-09T14:04:17.5162936Z configfs 57344 1 2025-09-09T14:04:17.5163176Z dmi_sysfs 20480 0 2025-09-09T14:04:17.5163410Z crc32_pclmul 16384 0 2025-09-09T14:04:17.5163657Z crc32c_intel 24576 0 2025-09-09T14:04:17.5164075Z efivarfs 24576 1 2025-09-09T14:04:17.5164308Z + modinfo nvidia 2025-09-09T14:04:17.5174516Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:04:17.5174978Z import_ns: DMA_BUF 2025-09-09T14:04:17.5175223Z alias: char-major-195-* 2025-09-09T14:04:17.5175473Z version: 580.65.06 2025-09-09T14:04:17.5175711Z supported: external 2025-09-09T14:04:17.5175949Z license: Dual MIT/GPL 2025-09-09T14:04:17.5176217Z firmware: nvidia/580.65.06/gsp_tu10x.bin 2025-09-09T14:04:17.5176536Z firmware: nvidia/580.65.06/gsp_ga10x.bin 2025-09-09T14:04:17.5176838Z srcversion: A69EBF72FC9D60E11E9A05C 2025-09-09T14:04:17.5177311Z alias: of:N*T*Cnvidia,tegra264-displayC* 2025-09-09T14:04:17.5177646Z alias: of:N*T*Cnvidia,tegra264-display 2025-09-09T14:04:17.5177980Z alias: of:N*T*Cnvidia,tegra234-displayC* 2025-09-09T14:04:17.5178318Z alias: of:N*T*Cnvidia,tegra234-display 2025-09-09T14:04:17.5178642Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:04:17.5178968Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:04:17.5179278Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:04:17.5179578Z depends: i2c-core,drm 2025-09-09T14:04:17.5179815Z retpoline: Y 2025-09-09T14:04:17.5180021Z name: nvidia 2025-09-09T14:04:17.5180362Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:04:17.5180819Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:04:17.5181245Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:04:17.5181652Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:04:17.5181960Z parm: NVreg_RmLogonRC:int 2025-09-09T14:04:17.5182245Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:04:17.5182553Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:04:17.5182842Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:04:17.5183138Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:04:17.5183479Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:04:17.5183855Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:04:17.5184168Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:04:17.5184457Z parm: NVreg_EnableMSI:int 2025-09-09T14:04:17.5184747Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:04:17.5185089Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:04:17.5185473Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:04:17.5185834Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:04:17.5186239Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:04:17.5186630Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:04:17.5187035Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:04:17.5187562Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:04:17.5187886Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:04:17.5188244Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:04:17.5188596Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:04:17.5188922Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:04:17.5189226Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:04:17.5189542Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:04:17.5189846Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:04:17.5190144Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:04:17.5190472Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:04:17.5190824Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:04:17.5191167Z parm: NVreg_RegisterPlatformDeviceDriver:int 2025-09-09T14:04:17.5191508Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:04:17.5191835Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:04:17.5192163Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:04:17.5192503Z parm: NVreg_CoherentGPUMemoryMode:charp 2025-09-09T14:04:17.5192823Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:04:17.5193147Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:04:17.5193472Z parm: NVreg_RmMsg:charp 2025-09-09T14:04:17.5193744Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:04:17.5194058Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:04:17.5194371Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:04:17.5194667Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:04:17.5194997Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:04:17.5195425Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:04:17.5195766Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:04:17.5196166Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:04:17.5196504Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:04:17.5196858Z parm: rm_firmware_active:charp 2025-09-09T14:04:17.5197128Z + set +e 2025-09-09T14:04:17.5197308Z + nvidia-smi 2025-09-09T14:04:19.8446961Z Tue Sep 9 14:04:19 2025 2025-09-09T14:04:19.8447315Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:19.8447792Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:04:19.8448250Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.8448727Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:04:19.8449253Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:04:19.8449666Z | | | MIG M. | 2025-09-09T14:04:19.8449990Z |=========================================+========================+======================| 2025-09-09T14:04:19.8787653Z | 0 NVIDIA A10G Off | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:04:19.8788525Z | 0% 28C P0 56W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:19.8789255Z | | | N/A | 2025-09-09T14:04:19.8789986Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.8790842Z | 1 NVIDIA A10G Off | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:04:19.8791665Z | 0% 29C P0 58W / 300W | 0MiB / 23028MiB | 2% Default | 2025-09-09T14:04:19.8792370Z | | | N/A | 2025-09-09T14:04:19.8793112Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.8794356Z | 2 NVIDIA A10G Off | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:04:19.8795159Z | 0% 29C P0 60W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:19.8795769Z | | | N/A | 2025-09-09T14:04:19.8796217Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.8796643Z | 3 NVIDIA A10G Off | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:04:19.8797042Z | 0% 28C P0 59W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:19.8797402Z | | | N/A | 2025-09-09T14:04:19.8797772Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:19.8798056Z 2025-09-09T14:04:19.8798220Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:19.8798640Z | Processes: | 2025-09-09T14:04:19.8799065Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:04:19.8799463Z | ID ID Usage | 2025-09-09T14:04:19.8799792Z |=========================================================================================| 2025-09-09T14:04:19.8815601Z | No running processes found | 2025-09-09T14:04:19.8816254Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:21.5443785Z + nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 2025-09-09T14:04:23.8555337Z NVIDIA A10G 2025-09-09T14:04:24.9501033Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:04:24.9501279Z + '[' 0 -eq 0 ']' 2025-09-09T14:04:24.9501510Z + echo 'INFO: Ignoring allowed status 0' 2025-09-09T14:04:24.9501782Z + set -e 2025-09-09T14:04:24.9501986Z INFO: Ignoring allowed status 0 2025-09-09T14:04:24.9512904Z == Installing nvidia container toolkit for amzn2023 == 2025-09-09T14:04:24.9518318Z + sudo yum install -y yum-utils 2025-09-09T14:04:25.3761060Z Last metadata expiration check: 0:07:25 ago on Tue Sep 9 13:57:00 2025. 2025-09-09T14:04:25.4007136Z Package dnf-utils-4.3.0-13.amzn2023.0.5.noarch is already installed. 2025-09-09T14:04:25.4469041Z Dependencies resolved. 2025-09-09T14:04:25.4698053Z Nothing to do. 2025-09-09T14:04:25.4698632Z Complete! 2025-09-09T14:04:25.5096407Z + [[ amzn2023 == \a\m\z\n\2\0\2\3 ]] 2025-09-09T14:04:25.5097058Z + YUM_REPO_URL=https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:04:25.5097890Z + sudo yum-config-manager --add-repo https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:04:25.8636375Z Adding repo from: https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:04:25.9115702Z + sudo yum install -y nvidia-container-toolkit-1.17.8 libnvidia-container-tools-1.17.8 libnvidia-container1-1.17.8 nvidia-container-toolkit-base-1.17.8 2025-09-09T14:04:26.4649090Z nvidia-container-toolkit 19 kB/s | 833 B 00:00 2025-09-09T14:04:26.4895456Z Package nvidia-container-toolkit-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.4901904Z Package libnvidia-container-tools-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.4905967Z Package libnvidia-container1-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.4913556Z Package nvidia-container-toolkit-base-1.17.8-1.x86_64 is already installed. 2025-09-09T14:04:26.5387860Z Dependencies resolved. 2025-09-09T14:04:26.5615640Z Nothing to do. 2025-09-09T14:04:26.5616298Z Complete! 2025-09-09T14:04:26.6021034Z + sudo systemctl restart docker 2025-09-09T14:04:44.7831919Z Tue Sep 9 14:04:44 2025 2025-09-09T14:04:44.7832299Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:44.7832784Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:04:44.7833257Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:44.7833735Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:04:44.7834248Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:04:44.7834681Z | | | MIG M. | 2025-09-09T14:04:44.7835001Z |=========================================+========================+======================| 2025-09-09T14:04:44.8183361Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:04:44.8184278Z | 0% 29C P0 56W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:44.8185009Z | | | N/A | 2025-09-09T14:04:44.8185840Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:44.8187223Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:04:44.8188207Z | 0% 29C P0 58W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:44.8189004Z | | | N/A | 2025-09-09T14:04:44.8190612Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:44.8191656Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:04:44.8192181Z | 0% 29C P0 60W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:04:44.8203119Z | | | N/A | 2025-09-09T14:04:44.8203520Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:44.8203951Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:04:44.8204368Z | 0% 28C P0 54W / 300W | 0MiB / 23028MiB | 2% Default | 2025-09-09T14:04:44.8204721Z | | | N/A | 2025-09-09T14:04:44.8205101Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:04:44.8205380Z 2025-09-09T14:04:44.8205542Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:44.8205960Z | Processes: | 2025-09-09T14:04:44.8206378Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:04:44.8206765Z | ID ID Usage | 2025-09-09T14:04:44.8207092Z |=========================================================================================| 2025-09-09T14:04:44.8212662Z | No running processes found | 2025-09-09T14:04:44.8213221Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:04:45.4469184Z Unable to find image 'public.ecr.aws/docker/library/python:3.13' locally 2025-09-09T14:04:45.7044087Z 3.13: Pulling from docker/library/python 2025-09-09T14:04:45.8083710Z 15b1d8a5ff03: Pulling fs layer 2025-09-09T14:04:45.8084080Z 22718812f617: Pulling fs layer 2025-09-09T14:04:45.8084724Z 401a98f7495b: Pulling fs layer 2025-09-09T14:04:45.8085056Z ad446e7df19a: Pulling fs layer 2025-09-09T14:04:45.8085371Z 5d32990caa16: Pulling fs layer 2025-09-09T14:04:45.8085684Z a79d633abf9a: Pulling fs layer 2025-09-09T14:04:45.8085944Z 249a56c8e466: Pulling fs layer 2025-09-09T14:04:45.8086182Z ad446e7df19a: Waiting 2025-09-09T14:04:45.8086401Z 5d32990caa16: Waiting 2025-09-09T14:04:45.8086610Z 249a56c8e466: Waiting 2025-09-09T14:04:45.8086831Z a79d633abf9a: Waiting 2025-09-09T14:04:45.9339444Z 22718812f617: Verifying Checksum 2025-09-09T14:04:45.9339734Z 22718812f617: Download complete 2025-09-09T14:04:45.9968538Z 15b1d8a5ff03: Download complete 2025-09-09T14:04:46.0501699Z 5d32990caa16: Verifying Checksum 2025-09-09T14:04:46.0502089Z 5d32990caa16: Download complete 2025-09-09T14:04:46.1672991Z 401a98f7495b: Verifying Checksum 2025-09-09T14:04:46.1673270Z 401a98f7495b: Download complete 2025-09-09T14:04:46.1676516Z a79d633abf9a: Verifying Checksum 2025-09-09T14:04:46.1676800Z a79d633abf9a: Download complete 2025-09-09T14:04:46.2341181Z 249a56c8e466: Download complete 2025-09-09T14:04:46.8085455Z ad446e7df19a: Verifying Checksum 2025-09-09T14:04:46.8085753Z ad446e7df19a: Download complete 2025-09-09T14:04:47.7611862Z 15b1d8a5ff03: Pull complete 2025-09-09T14:04:48.4545366Z 22718812f617: Pull complete 2025-09-09T14:04:50.9305522Z 401a98f7495b: Pull complete 2025-09-09T14:04:57.8789958Z ad446e7df19a: Pull complete 2025-09-09T14:04:58.3150684Z 5d32990caa16: Pull complete 2025-09-09T14:04:59.0625747Z a79d633abf9a: Pull complete 2025-09-09T14:04:59.0879403Z 249a56c8e466: Pull complete 2025-09-09T14:04:59.1020035Z Digest: sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T14:04:59.1065836Z Status: Downloaded newer image for public.ecr.aws/docker/library/python:3.13 2025-09-09T14:05:05.0511334Z Tue Sep 9 14:05:05 2025 2025-09-09T14:05:05.0511818Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:05:05.0512324Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:05:05.0512799Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:05.0513275Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:05:05.0513795Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:05:05.0514207Z | | | MIG M. | 2025-09-09T14:05:05.0514529Z |=========================================+========================+======================| 2025-09-09T14:05:05.1148224Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:05:05.1148656Z | 0% 26C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:05.1149024Z | | | N/A | 2025-09-09T14:05:05.1149403Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:05.1151964Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:05:05.1152492Z | 0% 26C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:05.1152846Z | | | N/A | 2025-09-09T14:05:05.1153220Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:05.1153633Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:05:05.1154043Z | 0% 26C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:05.1154404Z | | | N/A | 2025-09-09T14:05:05.1154989Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:05.1155411Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:05:05.1155809Z | 0% 25C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:05:05.1156229Z | | | N/A | 2025-09-09T14:05:05.1156593Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:05:05.1176443Z 2025-09-09T14:05:05.1176829Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:05:05.1177283Z | Processes: | 2025-09-09T14:05:05.1177718Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:05:05.1178130Z | ID ID Usage | 2025-09-09T14:05:05.1178467Z |=========================================================================================| 2025-09-09T14:05:05.1214349Z | No running processes found | 2025-09-09T14:05:05.1214840Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:05:07.0673200Z Command completed after 1 attempt(s). 2025-09-09T14:05:07.0761890Z ##[group]Run # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T14:05:07.0762417Z # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T14:05:07.0762814Z # shellcheck disable=SC2046 2025-09-09T14:05:07.0763322Z docker stop $(docker ps -q) || true 2025-09-09T14:05:07.0763636Z # Prune all of the docker images 2025-09-09T14:05:07.0764213Z docker system prune -af 2025-09-09T14:05:07.0779532Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:07.0779867Z env: 2025-09-09T14:05:07.0780110Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:07.0780431Z REPOSITORY: pytorch/ao 2025-09-09T14:05:07.0780674Z PR_NUMBER: 2963 2025-09-09T14:05:07.0781976Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:07.0783437Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:07.0783981Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:07.0784483Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:07.0784896Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:07.0785236Z ##[endgroup] 2025-09-09T14:05:07.1108345Z "docker stop" requires at least 1 argument. 2025-09-09T14:05:07.1108676Z See 'docker stop --help'. 2025-09-09T14:05:07.1108838Z 2025-09-09T14:05:07.1108984Z Usage: docker stop [OPTIONS] CONTAINER [CONTAINER...] 2025-09-09T14:05:07.1109230Z 2025-09-09T14:05:07.1109330Z Stop one or more running containers 2025-09-09T14:05:08.2901262Z Deleted Images: 2025-09-09T14:05:08.2902071Z untagged: public.ecr.aws/docker/library/python:3.13 2025-09-09T14:05:08.2902774Z untagged: public.ecr.aws/docker/library/python@sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T14:05:08.2903511Z deleted: sha256:77f2b24be2b3987f6d59918787d226acb4e6612644bacb3dd37adc494e477d9e 2025-09-09T14:05:08.2904113Z deleted: sha256:1b9aa91044866f8707424c8fe367f924a48557eac69f7485fd6d2a3a116c74d5 2025-09-09T14:05:08.2904682Z deleted: sha256:b86402d18e73d4825a3bd2a09244a93487ba4687ca7c9dcba0f73e160840845c 2025-09-09T14:05:08.2905579Z deleted: sha256:5755f8963eb047a0086073c3a7dd0731296d6751a7445f3693a52b30020a5b65 2025-09-09T14:05:08.2906146Z deleted: sha256:7f33dbfa9475d25622f49ed51f4164c97de1303331c77dfdc738e084d100f50c 2025-09-09T14:05:08.2906704Z deleted: sha256:19daa38049795ba2c166dd898c81b17e31f4b5f98c1337846c6515fff97d8782 2025-09-09T14:05:08.2907294Z deleted: sha256:483bd23b5d7e66fc0f8a92dbfacc3d72fad97ef47dd4767889979a803bc1f5b8 2025-09-09T14:05:08.2907875Z deleted: sha256:185e04da9d947141fd703dbf36361bdc2ff77cc27cbf500fb9f4881cb5ddbe95 2025-09-09T14:05:08.2908223Z 2025-09-09T14:05:08.3892381Z Total reclaimed space: 1.109GB 2025-09-09T14:05:08.3980362Z ##[group]Run ./test-infra/.github/actions/setup-ssh 2025-09-09T14:05:08.3980694Z with: 2025-09-09T14:05:08.3981336Z github-secret: *** 2025-09-09T14:05:08.3981970Z instructions: All testing is done inside the container, to start an interactive session run: docker exec -it $(docker container ps --format '{{.ID}}') bash 2025-09-09T14:05:08.3982662Z activate-with-label: false 2025-09-09T14:05:08.3982920Z label: with-ssh 2025-09-09T14:05:08.3983139Z remove-existing-keys: true 2025-09-09T14:05:08.3983403Z fail-silently: true 2025-09-09T14:05:08.3983618Z env: 2025-09-09T14:05:08.3983860Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:08.3984190Z REPOSITORY: pytorch/ao 2025-09-09T14:05:08.3984424Z PR_NUMBER: 2963 2025-09-09T14:05:08.3985738Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:08.3987202Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:08.3987742Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:08.3988255Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:08.3988669Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:08.3988992Z ##[endgroup] 2025-09-09T14:05:08.5097253Z Please see https://github.com/pytorch/pytorch/wiki/Debugging-using-with-ssh-for-Github-Actions for more info. 2025-09-09T14:05:09.0029172Z Grabbing public ssh keys from https://github.com/andrewor14.keys 2025-09-09T14:05:09.0775306Z ~/.ssh/authorized_keys file found on node, removing ~/.ssh and starting fresh 2025-09-09T14:05:09.0789521Z Public keys pulled and installed to /home/ec2-user/.ssh/authorized_keys 2025-09-09T14:05:09.0827613Z Login using: ssh ec2-user@ec2-54-174-250-109.compute-1.amazonaws.com 2025-09-09T14:05:09.0828188Z All testing is done inside the container, to start an interactive session run: 2025-09-09T14:05:09.0828683Z docker exec -it $(docker container ps --format '{{.ID}}') bash 2025-09-09T14:05:09.0983023Z ##[group]Run actions/checkout@11bd71901bbe5b1630ceea73d27597364c9af683 2025-09-09T14:05:09.0983429Z with: 2025-09-09T14:05:09.0983639Z repository: pytorch/ao 2025-09-09T14:05:09.0983882Z ref: refs/pull/2963/merge 2025-09-09T14:05:09.0984132Z path: pytorch/ao 2025-09-09T14:05:09.0984358Z fetch-depth: 1 2025-09-09T14:05:09.0984572Z submodules: recursive 2025-09-09T14:05:09.0984923Z token: *** 2025-09-09T14:05:09.0985125Z ssh-strict: true 2025-09-09T14:05:09.0985340Z ssh-user: git 2025-09-09T14:05:09.0985556Z persist-credentials: true 2025-09-09T14:05:09.0985808Z clean: true 2025-09-09T14:05:09.0986035Z sparse-checkout-cone-mode: true 2025-09-09T14:05:09.0986316Z fetch-tags: false 2025-09-09T14:05:09.0986543Z show-progress: true 2025-09-09T14:05:09.0986762Z lfs: false 2025-09-09T14:05:09.0986984Z set-safe-directory: true 2025-09-09T14:05:09.0987218Z env: 2025-09-09T14:05:09.0987460Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:09.0987824Z REPOSITORY: pytorch/ao 2025-09-09T14:05:09.0988061Z PR_NUMBER: 2963 2025-09-09T14:05:09.0989539Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:09.0990998Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:09.0991525Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:09.0992028Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:09.0992444Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:09.0992765Z ##[endgroup] 2025-09-09T14:05:09.1981308Z Syncing repository: pytorch/ao 2025-09-09T14:05:09.1989973Z ##[group]Getting Git version info 2025-09-09T14:05:09.1990393Z Working directory is '/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao' 2025-09-09T14:05:09.2016962Z [command]/usr/bin/git version 2025-09-09T14:05:09.2070002Z git version 2.47.1 2025-09-09T14:05:09.2095877Z ##[endgroup] 2025-09-09T14:05:09.2118816Z Temporarily overriding HOME='/home/ec2-user/actions-runner/_work/_temp/533a64ce-2cc0-4507-a1e0-95482da52235' before making global git config changes 2025-09-09T14:05:09.2119669Z Adding repository directory to the temporary git global config as a safe directory 2025-09-09T14:05:09.2124061Z [command]/usr/bin/git config --global --add safe.directory /home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao 2025-09-09T14:05:09.2164155Z ##[group]Initializing the repository 2025-09-09T14:05:09.2168129Z [command]/usr/bin/git init /home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao 2025-09-09T14:05:09.2214663Z hint: Using 'master' as the name for the initial branch. This default branch name 2025-09-09T14:05:09.2215256Z hint: is subject to change. To configure the initial branch name to use in all 2025-09-09T14:05:09.2215966Z hint: of your new repositories, which will suppress this warning, call: 2025-09-09T14:05:09.2216465Z hint: 2025-09-09T14:05:09.2216786Z hint: git config --global init.defaultBranch 2025-09-09T14:05:09.2217190Z hint: 2025-09-09T14:05:09.2217527Z hint: Names commonly chosen instead of 'master' are 'main', 'trunk' and 2025-09-09T14:05:09.2218035Z hint: 'development'. The just-created branch can be renamed via this command: 2025-09-09T14:05:09.2218435Z hint: 2025-09-09T14:05:09.2218634Z hint: git branch -m 2025-09-09T14:05:09.2219092Z Initialized empty Git repository in /home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao/.git/ 2025-09-09T14:05:09.2227179Z [command]/usr/bin/git remote add origin https://github.com/pytorch/ao 2025-09-09T14:05:09.2266889Z ##[endgroup] 2025-09-09T14:05:09.2267420Z ##[group]Disabling automatic garbage collection 2025-09-09T14:05:09.2303206Z [command]/usr/bin/git config --local gc.auto 0 2025-09-09T14:05:09.2308274Z ##[endgroup] 2025-09-09T14:05:09.2308645Z ##[group]Setting up auth 2025-09-09T14:05:09.2314099Z [command]/usr/bin/git config --local --name-only --get-regexp core\.sshCommand 2025-09-09T14:05:09.2349183Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'core\.sshCommand' && git config --local --unset-all 'core.sshCommand' || :" 2025-09-09T14:05:09.2761563Z [command]/usr/bin/git config --local --name-only --get-regexp http\.https\:\/\/github\.com\/\.extraheader 2025-09-09T14:05:09.2797923Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'http\.https\:\/\/github\.com\/\.extraheader' && git config --local --unset-all 'http.https://github.com/.extraheader' || :" 2025-09-09T14:05:09.3206190Z [command]/usr/bin/git config --local http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:05:09.3252347Z ##[endgroup] 2025-09-09T14:05:09.3252882Z ##[group]Fetching the repository 2025-09-09T14:05:09.3270322Z [command]/usr/bin/git -c protocol.version=2 fetch --no-tags --prune --no-recurse-submodules --depth=1 origin +refs/pull/2963/merge:refs/remotes/pull/2963/merge 2025-09-09T14:05:10.0930009Z From https://github.com/pytorch/ao 2025-09-09T14:05:10.0930497Z * [new ref] refs/pull/2963/merge -> pull/2963/merge 2025-09-09T14:05:10.0960195Z ##[endgroup] 2025-09-09T14:05:10.0960673Z ##[group]Determining the checkout info 2025-09-09T14:05:10.0962276Z ##[endgroup] 2025-09-09T14:05:10.0967308Z [command]/usr/bin/git sparse-checkout disable 2025-09-09T14:05:10.1009900Z [command]/usr/bin/git config --local --unset-all extensions.worktreeConfig 2025-09-09T14:05:10.1042426Z ##[group]Checking out the ref 2025-09-09T14:05:10.1045824Z [command]/usr/bin/git checkout --progress --force refs/remotes/pull/2963/merge 2025-09-09T14:05:10.2409832Z Note: switching to 'refs/remotes/pull/2963/merge'. 2025-09-09T14:05:10.2410120Z 2025-09-09T14:05:10.2410420Z You are in 'detached HEAD' state. You can look around, make experimental 2025-09-09T14:05:10.2410931Z changes and commit them, and you can discard any commits you make in this 2025-09-09T14:05:10.2411434Z state without impacting any branches by switching back to a branch. 2025-09-09T14:05:10.2411726Z 2025-09-09T14:05:10.2411922Z If you want to create a new branch to retain commits you create, you may 2025-09-09T14:05:10.2412382Z do so (now or later) by using -c with the switch command. Example: 2025-09-09T14:05:10.2412641Z 2025-09-09T14:05:10.2412748Z git switch -c 2025-09-09T14:05:10.2412939Z 2025-09-09T14:05:10.2413039Z Or undo this operation with: 2025-09-09T14:05:10.2413205Z 2025-09-09T14:05:10.2413298Z git switch - 2025-09-09T14:05:10.2413419Z 2025-09-09T14:05:10.2413635Z Turn off this advice by setting config variable advice.detachedHead to false 2025-09-09T14:05:10.2413955Z 2025-09-09T14:05:10.2414318Z HEAD is now at 7c05f81 Merge c21284c127b039bc49cc7ffda0e692894ed3b094 into 8b72284fd363b5c096de93fb7ac9cc960a6a601e 2025-09-09T14:05:10.2429100Z ##[endgroup] 2025-09-09T14:05:10.2429490Z ##[group]Setting up auth for fetching submodules 2025-09-09T14:05:10.2435226Z [command]/usr/bin/git config --global http.https://github.com/.extraheader AUTHORIZATION: basic *** 2025-09-09T14:05:10.2485560Z [command]/usr/bin/git config --global --unset-all url.https://github.com/.insteadOf 2025-09-09T14:05:10.2518975Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf git@github.com: 2025-09-09T14:05:10.2554750Z [command]/usr/bin/git config --global --add url.https://github.com/.insteadOf org-21003710@github.com: 2025-09-09T14:05:10.2585378Z ##[endgroup] 2025-09-09T14:05:10.2585876Z ##[group]Fetching submodules 2025-09-09T14:05:10.2589492Z [command]/usr/bin/git submodule sync --recursive 2025-09-09T14:05:10.2997223Z [command]/usr/bin/git -c protocol.version=2 submodule update --init --force --depth=1 --recursive 2025-09-09T14:05:10.3404657Z Submodule 'third_party/cutlass' (https://github.com/NVIDIA/cutlass) registered for path 'third_party/cutlass' 2025-09-09T14:05:10.3440790Z Cloning into '/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao/third_party/cutlass'... 2025-09-09T14:05:12.0654042Z From https://github.com/NVIDIA/cutlass 2025-09-09T14:05:12.0654542Z * branch e51efbfe18fe4f4cbb66ab814c55bf4aa0185491 -> FETCH_HEAD 2025-09-09T14:05:12.8416874Z Submodule path 'third_party/cutlass': checked out 'e51efbfe18fe4f4cbb66ab814c55bf4aa0185491' 2025-09-09T14:05:12.8468328Z [command]/usr/bin/git submodule foreach --recursive git config --local gc.auto 0 2025-09-09T14:05:12.8861868Z Entering 'third_party/cutlass' 2025-09-09T14:05:12.8947238Z ##[endgroup] 2025-09-09T14:05:12.8947637Z ##[group]Persisting credentials for submodules 2025-09-09T14:05:12.8953527Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'url\.https\:\/\/github\.com\/\.insteadOf' && git config --local --unset-all 'url.https://github.com/.insteadOf' || :" 2025-09-09T14:05:12.9333902Z Entering 'third_party/cutlass' 2025-09-09T14:05:12.9443195Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local 'http.https://github.com/.extraheader' 'AUTHORIZATION: basic ***' && git config --local --show-origin --name-only --get-regexp remote.origin.url" 2025-09-09T14:05:12.9829925Z Entering 'third_party/cutlass' 2025-09-09T14:05:12.9905180Z file:/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao/.git/modules/third_party/cutlass/config remote.origin.url 2025-09-09T14:05:12.9974236Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'git@github.com:' 2025-09-09T14:05:13.0365161Z Entering 'third_party/cutlass' 2025-09-09T14:05:13.0451383Z [command]/usr/bin/git submodule foreach --recursive git config --local --add 'url.https://github.com/.insteadOf' 'org-21003710@github.com:' 2025-09-09T14:05:13.0840410Z Entering 'third_party/cutlass' 2025-09-09T14:05:13.0921539Z ##[endgroup] 2025-09-09T14:05:13.0966700Z [command]/usr/bin/git log -1 --format=%H 2025-09-09T14:05:13.0996846Z 7c05f811b89289f7be3e0e3546626827f2cc1ca4 2025-09-09T14:05:13.1212485Z Prepare all required actions 2025-09-09T14:05:13.1213295Z Getting action download info 2025-09-09T14:05:13.2909904Z Download action repository 'nick-fields/retry@v3.0.0' (SHA:7152eba30c6575329ac0576536151aca5a72780e) 2025-09-09T14:05:13.4639194Z ##[group]Run ./test-infra/.github/actions/calculate-docker-image 2025-09-09T14:05:13.4639573Z with: 2025-09-09T14:05:13.4639810Z use-custom-docker-registry: true 2025-09-09T14:05:13.4640178Z docker-image-name: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:13.4640530Z docker-build-dir: .ci/docker 2025-09-09T14:05:13.4640818Z working-directory: pytorch/ao 2025-09-09T14:05:13.4641098Z docker-build-script: ./build.sh 2025-09-09T14:05:13.4641468Z docker-registry: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:13.4641839Z force-push: false 2025-09-09T14:05:13.4642064Z env: 2025-09-09T14:05:13.4642315Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:13.4642672Z REPOSITORY: pytorch/ao 2025-09-09T14:05:13.4642954Z PR_NUMBER: 2963 2025-09-09T14:05:13.4644282Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:13.4645758Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:13.4646306Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:13.4646822Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:13.4647240Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:13.4647582Z ##[endgroup] 2025-09-09T14:05:13.4671835Z ##[group]Run set -ex 2025-09-09T14:05:13.4672121Z set -ex 2025-09-09T14:05:13.4672329Z  2025-09-09T14:05:13.4672706Z # If the docker build directory or the build script doesn't exist, the action will 2025-09-09T14:05:13.4673301Z # gracefully return the docker image name as it is. Pulling docker image in Linux 2025-09-09T14:05:13.4673820Z # job could then download the pre-built image as usual 2025-09-09T14:05:13.4674448Z if [[ -d "${DOCKER_BUILD_DIR}" ]] && [[ -f "${DOCKER_BUILD_DIR}/${DOCKER_BUILD_SCRIPT}" ]] && [[ "${USE_CUSTOM_DOCKER_REGISTRY}" == "true" ]]; then 2025-09-09T14:05:13.4675017Z  echo "skip=false" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4675314Z else 2025-09-09T14:05:13.4675551Z  echo "skip=true" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4675973Z  echo "docker-image=${DOCKER_IMAGE_NAME}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4676421Z  2025-09-09T14:05:13.4676916Z  echo "Not using custom ECR registry. Either it was not requested or there is no Docker build script in the ${REPO_NAME} repo..." 2025-09-09T14:05:13.4677655Z  exit 0 2025-09-09T14:05:13.4677853Z fi 2025-09-09T14:05:13.4678051Z  2025-09-09T14:05:13.4678355Z if [[ "${DOCKER_IMAGE_NAME}" == *"${DOCKER_REGISTRY}/${REPO_NAME}"* ]]; then 2025-09-09T14:05:13.4678895Z  # The docker image name already includes the ECR prefix and tag, so we can just 2025-09-09T14:05:13.4679370Z  # use it as it is, but first let's extract the tag 2025-09-09T14:05:13.4679797Z  DOCKER_TAG=$(echo "${DOCKER_IMAGE_NAME}" | awk -F '[:,]' '{print $2}') 2025-09-09T14:05:13.4680266Z  echo "docker-tag=${DOCKER_TAG}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4680696Z  echo "docker-image=${DOCKER_IMAGE_NAME}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4681064Z else 2025-09-09T14:05:13.4681307Z  if [[ "${DOCKER_IMAGE_NAME}" == *:* ]]; then 2025-09-09T14:05:13.4681651Z  CUSTOM_TAG_PREFIX=${DOCKER_IMAGE_NAME#*:} 2025-09-09T14:05:13.4682206Z  DOCKER_IMAGE_NAME=${DOCKER_IMAGE_NAME%%:*} 2025-09-09T14:05:13.4682502Z  fi 2025-09-09T14:05:13.4682913Z  DOCKER_TAG=${CUSTOM_TAG_PREFIX:+${CUSTOM_TAG_PREFIX}-}$(git rev-parse HEAD:"${DOCKER_BUILD_DIR}") 2025-09-09T14:05:13.4683460Z  echo "docker-tag=${DOCKER_TAG}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4684050Z  echo "docker-image=${DOCKER_REGISTRY}/${REPO_NAME}/${DOCKER_IMAGE_NAME}:${DOCKER_TAG}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4684693Z  echo "custom-tag-prefix=${CUSTOM_TAG_PREFIX}" >> "${GITHUB_OUTPUT}" 2025-09-09T14:05:13.4685079Z fi 2025-09-09T14:05:13.4694236Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:13.4694582Z env: 2025-09-09T14:05:13.4694835Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:13.4695161Z REPOSITORY: pytorch/ao 2025-09-09T14:05:13.4695414Z PR_NUMBER: 2963 2025-09-09T14:05:13.4696739Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:13.4698216Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:13.4698752Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:13.4699250Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:13.4699679Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:13.4700006Z REPO_NAME: ao 2025-09-09T14:05:13.4700307Z DOCKER_IMAGE_NAME: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:13.4700684Z DOCKER_BUILD_DIR: .ci/docker 2025-09-09T14:05:13.4700956Z DOCKER_BUILD_SCRIPT: ./build.sh 2025-09-09T14:05:13.4701320Z DOCKER_REGISTRY: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:13.4701690Z USE_CUSTOM_DOCKER_REGISTRY: true 2025-09-09T14:05:13.4701968Z CUSTOM_TAG_PREFIX: 2025-09-09T14:05:13.4702188Z ##[endgroup] 2025-09-09T14:05:13.4734393Z + [[ -d .ci/docker ]] 2025-09-09T14:05:13.4734626Z + echo skip=true 2025-09-09T14:05:13.4734904Z + echo docker-image=pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:13.4735498Z + echo 'Not using custom ECR registry. Either it was not requested or there is no Docker build script in the ao repo...' 2025-09-09T14:05:13.4736010Z + exit 0 2025-09-09T14:05:13.4736424Z Not using custom ECR registry. Either it was not requested or there is no Docker build script in the ao repo... 2025-09-09T14:05:13.4789648Z ##[group]Run set -eux 2025-09-09T14:05:13.4789929Z set -eux 2025-09-09T14:05:13.4790316Z # It's ok if this steps fails, it would then be an anonymous user like what we used to have 2025-09-09T14:05:13.4791579Z aws secretsmanager get-secret-value --secret-id docker_hub_readonly_token | jq --raw-output '.SecretString' | jq -r .docker_hub_readonly_token | docker login --username pytorchbot --password-stdin || true 2025-09-09T14:05:13.4801459Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:13.4801819Z env: 2025-09-09T14:05:13.4802075Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:13.4802431Z REPOSITORY: pytorch/ao 2025-09-09T14:05:13.4802681Z PR_NUMBER: 2963 2025-09-09T14:05:13.4804000Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:13.4805463Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:13.4806194Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:13.4806695Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:13.4807122Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:13.4807449Z ##[endgroup] 2025-09-09T14:05:13.4844694Z + aws secretsmanager get-secret-value --secret-id docker_hub_readonly_token 2025-09-09T14:05:13.4845681Z + jq --raw-output .SecretString 2025-09-09T14:05:13.4847441Z + jq -r .docker_hub_readonly_token 2025-09-09T14:05:13.4849008Z + docker login --username pytorchbot --password-stdin 2025-09-09T14:05:14.0870185Z WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json. 2025-09-09T14:05:14.0870870Z Configure a credential helper to remove this warning. See 2025-09-09T14:05:14.0871397Z https://docs.docker.com/engine/reference/commandline/login/#credentials-store 2025-09-09T14:05:14.0871760Z 2025-09-09T14:05:14.0871851Z Login Succeeded 2025-09-09T14:05:14.0960750Z Prepare all required actions 2025-09-09T14:05:14.0996663Z ##[group]Run ./test-infra/.github/actions/pull-docker-image 2025-09-09T14:05:14.0997010Z with: 2025-09-09T14:05:14.0997258Z docker-image: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.0997680Z docker-registry: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:14.0998032Z env: 2025-09-09T14:05:14.0998284Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.0998603Z REPOSITORY: pytorch/ao 2025-09-09T14:05:14.0998852Z PR_NUMBER: 2963 2025-09-09T14:05:14.1000152Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:14.1001635Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:14.1002169Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:14.1002670Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:14.1003089Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:14.1003412Z ##[endgroup] 2025-09-09T14:05:14.1023101Z ##[group]Run set -x 2025-09-09T14:05:14.1023360Z set -x 2025-09-09T14:05:14.1023569Z set +e 2025-09-09T14:05:14.1023761Z  2025-09-09T14:05:14.1023963Z login() { 2025-09-09T14:05:14.1024392Z  aws ecr get-login-password --region us-east-1 | docker login -u AWS --password-stdin "$1" 2025-09-09T14:05:14.1024871Z } 2025-09-09T14:05:14.1025057Z  2025-09-09T14:05:14.1025249Z retry () { 2025-09-09T14:05:14.1025499Z  $* || (sleep 1 && $*) || (sleep 2 && $*) 2025-09-09T14:05:14.1025779Z } 2025-09-09T14:05:14.1026152Z  2025-09-09T14:05:14.1026363Z retry login "${DOCKER_REGISTRY}" 2025-09-09T14:05:14.1026640Z  2025-09-09T14:05:14.1027065Z IMAGE_SIZE=$(docker manifest inspect "${DOCKER_IMAGE}" | jq '[.layers[].size, .config.size] | add / 1024 / 1024') 2025-09-09T14:05:14.1027658Z echo "Compressed size of image in MB: ${IMAGE_SIZE}" 2025-09-09T14:05:14.1027991Z  2025-09-09T14:05:14.1028186Z set -e 2025-09-09T14:05:14.1028505Z # ignore output since only exit code is used for conditional 2025-09-09T14:05:14.1028945Z # only pull docker image if it's not available locally 2025-09-09T14:05:14.1029439Z if ! docker inspect --type=image "${DOCKER_IMAGE}" >/dev/null 2>/dev/null; then 2025-09-09T14:05:14.1029886Z  retry docker pull "${DOCKER_IMAGE}" 2025-09-09T14:05:14.1030181Z fi 2025-09-09T14:05:14.1039684Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:05:14.1040044Z env: 2025-09-09T14:05:14.1040302Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.1040646Z REPOSITORY: pytorch/ao 2025-09-09T14:05:14.1040892Z PR_NUMBER: 2963 2025-09-09T14:05:14.1042196Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:05:14.1043666Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:05:14.1044210Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:05:14.1044708Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:05:14.1045133Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:05:14.1045728Z DOCKER_REGISTRY: 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:14.1046102Z ##[endgroup] 2025-09-09T14:05:14.1081486Z + set +e 2025-09-09T14:05:14.1082015Z + retry login 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:14.1082412Z + login 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:14.1086732Z + aws ecr get-login-password --region us-east-1 2025-09-09T14:05:14.1087532Z + docker login -u AWS --password-stdin 308535385114.dkr.ecr.us-east-1.amazonaws.com 2025-09-09T14:05:14.6590450Z WARNING! Your password will be stored unencrypted in /home/ec2-user/.docker/config.json. 2025-09-09T14:05:14.6591348Z Configure a credential helper to remove this warning. See 2025-09-09T14:05:14.6591916Z https://docs.docker.com/engine/reference/commandline/login/#credentials-store 2025-09-09T14:05:14.6592266Z 2025-09-09T14:05:14.6592354Z Login Succeeded 2025-09-09T14:05:14.6626176Z ++ docker manifest inspect pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.6627067Z ++ jq '[.layers[].size, .config.size] | add / 1024 / 1024' 2025-09-09T14:05:14.8263621Z + IMAGE_SIZE=7943.772059440613 2025-09-09T14:05:14.8264122Z + echo 'Compressed size of image in MB: 7943.772059440613' 2025-09-09T14:05:14.8264450Z + set -e 2025-09-09T14:05:14.8264770Z + docker inspect --type=image pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.8265192Z Compressed size of image in MB: 7943.772059440613 2025-09-09T14:05:14.8432157Z + retry docker pull pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:14.8432561Z + docker pull pytorch/almalinux-builder:cuda12.6 2025-09-09T14:05:15.0153152Z cuda12.6: Pulling from pytorch/almalinux-builder 2025-09-09T14:05:15.0153698Z 19877a9af8e3: Pulling fs layer 2025-09-09T14:05:15.0154091Z 3b95f7accc18: Pulling fs layer 2025-09-09T14:05:15.0154460Z 09fcdf4cf4fd: Pulling fs layer 2025-09-09T14:05:15.0154823Z 17af5086235f: Pulling fs layer 2025-09-09T14:05:15.0155173Z c3175a707c2d: Pulling fs layer 2025-09-09T14:05:15.0155531Z 550b3c83242f: Pulling fs layer 2025-09-09T14:05:15.0155914Z 018f40a634ae: Pulling fs layer 2025-09-09T14:05:15.0156579Z 4f4fb700ef54: Pulling fs layer 2025-09-09T14:05:15.0156825Z cabce7a916a3: Pulling fs layer 2025-09-09T14:05:15.0157074Z 0b3a66ab554e: Pulling fs layer 2025-09-09T14:05:15.0157316Z 72728e4acc07: Pulling fs layer 2025-09-09T14:05:15.0157565Z 2ca30f8660e0: Pulling fs layer 2025-09-09T14:05:15.0157816Z 45f90a05dbb6: Pulling fs layer 2025-09-09T14:05:15.0158060Z 16125e2dbaa8: Pulling fs layer 2025-09-09T14:05:15.0158306Z 8e08c86db3a1: Pulling fs layer 2025-09-09T14:05:15.0158551Z 550d67135f81: Pulling fs layer 2025-09-09T14:05:15.0158799Z cac5e14b36bd: Pulling fs layer 2025-09-09T14:05:15.0159044Z d9fc50fb0d36: Pulling fs layer 2025-09-09T14:05:15.0159294Z 8b2ffa49399c: Pulling fs layer 2025-09-09T14:05:15.0159536Z 05e4f4570fa0: Pulling fs layer 2025-09-09T14:05:15.0159785Z e16b313a64bb: Pulling fs layer 2025-09-09T14:05:15.0160026Z 87b47e27ca53: Pulling fs layer 2025-09-09T14:05:15.0160285Z 282dc51a39ad: Pulling fs layer 2025-09-09T14:05:15.0160525Z 550b3c83242f: Waiting 2025-09-09T14:05:15.0160743Z 018f40a634ae: Waiting 2025-09-09T14:05:15.0160949Z 4f4fb700ef54: Waiting 2025-09-09T14:05:15.0161155Z cabce7a916a3: Waiting 2025-09-09T14:05:15.0161367Z 17af5086235f: Waiting 2025-09-09T14:05:15.0161569Z 0b3a66ab554e: Waiting 2025-09-09T14:05:15.0161781Z d9fc50fb0d36: Waiting 2025-09-09T14:05:15.0161988Z 72728e4acc07: Waiting 2025-09-09T14:05:15.0162206Z 87b47e27ca53: Waiting 2025-09-09T14:05:15.0162411Z 8b2ffa49399c: Waiting 2025-09-09T14:05:15.0162620Z 282dc51a39ad: Waiting 2025-09-09T14:05:15.0162831Z 05e4f4570fa0: Waiting 2025-09-09T14:05:15.0163035Z e16b313a64bb: Waiting 2025-09-09T14:05:15.0163248Z 2ca30f8660e0: Waiting 2025-09-09T14:05:15.0163455Z cac5e14b36bd: Waiting 2025-09-09T14:05:15.0163667Z 45f90a05dbb6: Waiting 2025-09-09T14:05:15.0164047Z c3175a707c2d: Waiting 2025-09-09T14:05:15.0164256Z 550d67135f81: Waiting 2025-09-09T14:05:15.0164458Z 8e08c86db3a1: Waiting 2025-09-09T14:05:15.0884803Z 09fcdf4cf4fd: Verifying Checksum 2025-09-09T14:05:15.0885868Z 09fcdf4cf4fd: Download complete 2025-09-09T14:05:15.5282158Z 17af5086235f: Verifying Checksum 2025-09-09T14:05:15.5282481Z 17af5086235f: Download complete 2025-09-09T14:05:15.7816928Z 19877a9af8e3: Verifying Checksum 2025-09-09T14:05:15.7817240Z 19877a9af8e3: Download complete 2025-09-09T14:05:15.8187671Z 550b3c83242f: Verifying Checksum 2025-09-09T14:05:15.8188133Z 550b3c83242f: Download complete 2025-09-09T14:05:16.2796842Z 3b95f7accc18: Verifying Checksum 2025-09-09T14:05:16.2797142Z 3b95f7accc18: Download complete 2025-09-09T14:05:16.2940726Z 018f40a634ae: Verifying Checksum 2025-09-09T14:05:16.2941013Z 018f40a634ae: Download complete 2025-09-09T14:05:16.3240509Z 4f4fb700ef54: Download complete 2025-09-09T14:05:16.3388841Z cabce7a916a3: Verifying Checksum 2025-09-09T14:05:16.3389375Z cabce7a916a3: Download complete 2025-09-09T14:05:16.3632247Z 0b3a66ab554e: Verifying Checksum 2025-09-09T14:05:16.3632537Z 0b3a66ab554e: Download complete 2025-09-09T14:05:16.3988156Z 2ca30f8660e0: Download complete 2025-09-09T14:05:16.4252794Z 72728e4acc07: Verifying Checksum 2025-09-09T14:05:16.4253174Z 72728e4acc07: Download complete 2025-09-09T14:05:16.5369824Z 16125e2dbaa8: Download complete 2025-09-09T14:05:16.5826113Z 8e08c86db3a1: Download complete 2025-09-09T14:05:17.4491505Z c3175a707c2d: Verifying Checksum 2025-09-09T14:05:17.4491865Z c3175a707c2d: Download complete 2025-09-09T14:05:17.4788024Z cac5e14b36bd: Verifying Checksum 2025-09-09T14:05:17.4788305Z cac5e14b36bd: Download complete 2025-09-09T14:05:17.5194598Z d9fc50fb0d36: Verifying Checksum 2025-09-09T14:05:17.5194871Z d9fc50fb0d36: Download complete 2025-09-09T14:05:17.5807230Z 8b2ffa49399c: Download complete 2025-09-09T14:05:17.7115365Z 05e4f4570fa0: Verifying Checksum 2025-09-09T14:05:17.7115915Z 05e4f4570fa0: Download complete 2025-09-09T14:05:17.7478290Z e16b313a64bb: Download complete 2025-09-09T14:05:17.7838564Z 87b47e27ca53: Verifying Checksum 2025-09-09T14:05:17.7838845Z 87b47e27ca53: Download complete 2025-09-09T14:05:18.4206333Z 19877a9af8e3: Pull complete 2025-09-09T14:05:20.8737066Z 3b95f7accc18: Pull complete 2025-09-09T14:05:20.9487969Z 550d67135f81: Verifying Checksum 2025-09-09T14:05:20.9488251Z 550d67135f81: Download complete 2025-09-09T14:05:21.0415344Z 09fcdf4cf4fd: Pull complete 2025-09-09T14:05:21.3171547Z 17af5086235f: Pull complete 2025-09-09T14:05:23.6283467Z 282dc51a39ad: Verifying Checksum 2025-09-09T14:05:23.6283824Z 282dc51a39ad: Download complete 2025-09-09T14:05:27.1751574Z c3175a707c2d: Pull complete 2025-09-09T14:05:27.1983242Z 550b3c83242f: Pull complete 2025-09-09T14:05:28.4578703Z 018f40a634ae: Pull complete 2025-09-09T14:05:28.4806080Z 4f4fb700ef54: Pull complete 2025-09-09T14:06:14.7394533Z cabce7a916a3: Pull complete 2025-09-09T14:06:14.7628090Z 0b3a66ab554e: Pull complete 2025-09-09T14:06:14.7869095Z 72728e4acc07: Pull complete 2025-09-09T14:06:14.8090485Z 2ca30f8660e0: Pull complete 2025-09-09T14:06:24.6204366Z 45f90a05dbb6: Verifying Checksum 2025-09-09T14:06:24.6204685Z 45f90a05dbb6: Download complete 2025-09-09T14:07:24.9586791Z 45f90a05dbb6: Pull complete 2025-09-09T14:07:25.4998400Z 16125e2dbaa8: Pull complete 2025-09-09T14:07:25.8826776Z 8e08c86db3a1: Pull complete 2025-09-09T14:07:41.6677260Z 550d67135f81: Pull complete 2025-09-09T14:07:42.1189137Z cac5e14b36bd: Pull complete 2025-09-09T14:07:42.5251151Z d9fc50fb0d36: Pull complete 2025-09-09T14:07:42.8709181Z 8b2ffa49399c: Pull complete 2025-09-09T14:07:43.4526217Z 05e4f4570fa0: Pull complete 2025-09-09T14:07:43.8087086Z e16b313a64bb: Pull complete 2025-09-09T14:07:44.1683042Z 87b47e27ca53: Pull complete 2025-09-09T14:08:03.2514390Z 282dc51a39ad: Pull complete 2025-09-09T14:08:03.4579978Z Digest: sha256:be7f2a4c6f467933b154ac0b3ded894ad1bf06ce95f8f8d908dba108e68806f3 2025-09-09T14:08:03.5610841Z Status: Downloaded newer image for pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:03.5950771Z docker.io/pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:03.6015864Z ##[group]Run echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:08:03.6016753Z echo "IN_CONTAINER_RUNNER=$(if [ -f /.inarc ] || [ -f /.incontainer ]; then echo true ; else echo false; fi)" >> "$GITHUB_OUTPUT" 2025-09-09T14:08:03.6028743Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T14:08:03.6029079Z env: 2025-09-09T14:08:03.6029314Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:03.6029635Z REPOSITORY: pytorch/ao 2025-09-09T14:08:03.6029862Z PR_NUMBER: 2963 2025-09-09T14:08:03.6031189Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:08:03.6032634Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:08:03.6033172Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:08:03.6033672Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:08:03.6034083Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:08:03.6034414Z ##[endgroup] 2025-09-09T14:08:03.6257993Z Prepare all required actions 2025-09-09T14:08:03.6258339Z Getting action download info 2025-09-09T14:08:03.8219168Z ##[group]Run ./test-infra/.github/actions/setup-nvidia 2025-09-09T14:08:03.8219511Z with: 2025-09-09T14:08:03.8219724Z driver-version: 580.65.06 2025-09-09T14:08:03.8219971Z env: 2025-09-09T14:08:03.8220212Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:03.8220553Z REPOSITORY: pytorch/ao 2025-09-09T14:08:03.8220795Z PR_NUMBER: 2963 2025-09-09T14:08:03.8222118Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:08:03.8223847Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:08:03.8224422Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:08:03.8224987Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:08:03.8225409Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:08:03.8225732Z ##[endgroup] 2025-09-09T14:08:03.8357879Z ##[group]Run nick-fields/retry@3e91a01664abd3c5cd539100d10d33b9c5b68482 2025-09-09T14:08:03.8358409Z with: 2025-09-09T14:08:03.8358668Z timeout_minutes: 10 2025-09-09T14:08:03.8358957Z max_attempts: 3 2025-09-09T14:08:03.8383057Z command: # Is it disgusting to have a full shell script here in this github action? Sure # But is it the best way to make it so that this action relies on nothing else? Absolutely set -eou pipefail DISTRIBUTION=$(. /etc/os-release;echo $ID$VERSION_ID) DRIVER_FN="NVIDIA-Linux-x86_64-${DRIVER_VERSION}.run" install_nvidia_docker2_amzn2() { ( set -x # Needed for yum-config-manager sudo yum install -y yum-utils if [[ "${DISTRIBUTION}" == "amzn2023" ]] ; then YUM_REPO_URL="https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo" else # Amazon Linux 2 YUM_REPO_URL="https://nvidia.github.io/nvidia-docker/${DISTRIBUTION}/nvidia-docker.repo" fi sudo yum-config-manager --add-repo "${YUM_REPO_URL}" sudo yum install -y \ nvidia-container-toolkit-1.17.8 \ libnvidia-container-tools-1.17.8 \ libnvidia-container1-1.17.8 \ nvidia-container-toolkit-base-1.17.8 sudo systemctl restart docker ) } install_nvidia_docker2_ubuntu20() { ( set -x # Install nvidia-driver package if not installed status="$(dpkg-query -W --showformat='${db:Status-Status}' nvidia-docker2 2>&1)" if [ ! $? = 0 ] || [ ! "$status" = installed ]; then sudo apt-get install -y nvidia-container-toolkit-1.17.8 sudo systemctl restart docker fi ) } pre_install_nvidia_driver_amzn2() { ( # Purge any nvidia driver installed from RHEL repo sudo yum remove -y nvidia-driver-latest-dkms ) } install_nvidia_driver_common() { ( # Try to gather more information about the runner and its existing NVIDIA driver if any echo "Before installing NVIDIA driver" lspci lsmod modinfo nvidia || true HAS_NVIDIA_DRIVER=0 # Check if NVIDIA driver has already been installed if [ -x "$(command -v nvidia-smi)" ]; then set +e # The driver exists, check its version next. Also check only the first GPU if there are more than one of them # so that the same driver version is not print over multiple lines INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then echo "Failed to get NVIDIA driver version ($INSTALLED_DRIVER_VERSION). Continuing" elif [ "$INSTALLED_DRIVER_VERSION" != "$DRIVER_VERSION" ]; then echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has been installed, but we expect to have $DRIVER_VERSION instead. Continuing" # Turn off persistent mode so that the installation script can unload the kernel module sudo killall nvidia-persistenced || true else HAS_NVIDIA_DRIVER=1 echo "NVIDIA driver ($INSTALLED_DRIVER_VERSION) has already been installed. Skipping NVIDIA driver installation" fi set -e fi if [ "$HAS_NVIDIA_DRIVER" -eq 0 ]; then # CAUTION: this may need to be updated in future if [ "${DISTRIBUTION}" != ubuntu20.04 ]; then sudo yum groupinstall -y "Development Tools" # ensure our kernel install is the same as our underlying kernel, # groupinstall "Development Tools" has a habit of mismatching kernel headers sudo yum install -y "kernel-devel-uname-r == $(uname -r)" sudo modprobe backlight fi sudo curl -fsL -o /tmp/nvidia_driver "https://s3.amazonaws.com/ossci-linux/nvidia_driver/$DRIVER_FN" set +e sudo /bin/bash /tmp/nvidia_driver -s --no-drm NVIDIA_INSTALLATION_STATUS=$? RESET_GPU=0 if [ "$NVIDIA_INSTALLATION_STATUS" -ne 0 ]; then sudo cat /var/log/nvidia-installer.log # Fail to install NVIDIA driver, try to reset the GPU RESET_GPU=1 elif [ -x "$(command -v nvidia-smi)" ]; then # Check again if nvidia-smi works even if the driver installation completes successfully INSTALLED_DRIVER_VERSION=$(nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0) NVIDIA_SMI_STATUS=$? if [ "$NVIDIA_SMI_STATUS" -ne 0 ] && [ "$NVIDIA_SMI_STATUS" -ne 14 ]; then RESET_GPU=1 fi fi if [ "$RESET_GPU" -eq 1 ]; then NVIDIA_DEVICES=$(lspci -D | grep -i NVIDIA | cut -d' ' -f1) # The GPU can get stuck in a failure state if somehow the test crashs the GPU microcode. When this # happens, we'll try to reset all NVIDIA devices https://github.com/pytorch/pytorch/issues/88388 for PCI_ID in $NVIDIA_DEVICES; do DEVICE_ENABLED=$(cat /sys/bus/pci/devices/$PCI_ID/enable) echo "Reseting $PCI_ID (enabled state: $DEVICE_ENABLED)" # This requires sudo permission of course echo "1" | sudo tee /sys/bus/pci/devices/$PCI_ID/reset sleep 1 done fi sudo rm -fv /tmp/nvidia_driver set -e fi ) } post_install_nvidia_driver_common() { ( sudo modprobe nvidia || true echo "After installing NVIDIA driver" lspci lsmod modinfo nvidia || true ( set +e nvidia-smi # NB: Annoyingly, nvidia-smi command returns successfully with return code 0 even in # the case where the driver has already crashed as it still can get the driver version # and some basic information like the bus ID. However, the rest of the information # would be missing (ERR!), for example: # # +-----------------------------------------------------------------------------+ # | NVIDIA-SMI 525.89.02 Driver Version: 525.89.02 CUDA Version: 12.0 | # |-------------------------------+----------------------+----------------------+ # | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | # | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | # | | | MIG M. | # |===============================+======================+======================| # | 0 ERR! Off | 00000000:00:1E.0 Off | ERR! | # |ERR! ERR! ERR! ERR! / ERR! | 4184MiB / 23028MiB | ERR! Default | # | | | ERR! | # +-------------------------------+----------------------+----------------------+ # # +-----------------------------------------------------------------------------+ # | Processes: | # | GPU GI CI PID Type Process name GPU Memory | # | ID ID Usage | # |=============================================================================| # +-----------------------------------------------------------------------------+ # # This should be reported as a failure instead as it will guarantee to fail when # Docker tries to run with --gpus all # # So, the correct check here is to query one of the missing piece of info like # GPU name, so that the command can fail accordingly nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 NVIDIA_SMI_STATUS=$? # Allowable exit statuses for nvidia-smi, see: https://github.com/NVIDIA/gpu-operator/issues/285 if [ "$NVIDIA_SMI_STATUS" -eq 0 ] || [ "$NVIDIA_SMI_STATUS" -eq 14 ]; then echo "INFO: Ignoring allowed status ${NVIDIA_SMI_STATUS}" else echo "ERROR: nvidia-smi exited with unresolved status ${NVIDIA_SMI_STATUS}" exit ${NVIDIA_SMI_STATUS} fi set -e ) ) } install_nvidia_driver_amzn2() { ( set -x pre_install_nvidia_driver_amzn2 install_nvidia_driver_common post_install_nvidia_driver_common ) } install_nvidia_driver_ubuntu20() { ( set -x install_nvidia_driver_common post_install_nvidia_driver_common ) } echo "== Installing nvidia driver ${DRIVER_FN} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_driver_amzn2 ;; ubuntu20.04) install_nvidia_driver_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac # Install container toolkit based on distribution echo "== Installing nvidia container toolkit for ${DISTRIBUTION} ==" case "${DISTRIBUTION}" in amzn*) install_nvidia_docker2_amzn2 ;; ubuntu20.04) install_nvidia_docker2_ubuntu20 ;; *) echo "ERROR: Unknown distribution ${DISTRIBUTION}" exit 1 ;; esac echo "GPU_FLAG=--gpus all -e NVIDIA_DRIVER_CAPABILITIES=all" >> "${GITHUB_ENV}" # Fix https://github.com/NVIDIA/nvidia-docker/issues/1648 on runners with # more than one GPUs. This just needs to be run once. The command fails # on subsequent runs and complains that the mode is already on, but that's # ok sudo nvidia-persistenced || true # This should show persistence mode ON nvidia-smi # check if the container-toolkit is correctly installed and CUDA is available inside a container docker run --rm -t --gpus=all public.ecr.aws/docker/library/python:3.13 nvidia-smi 2025-09-09T14:08:03.8408051Z retry_wait_seconds: 10 2025-09-09T14:08:03.8408302Z polling_interval_seconds: 1 2025-09-09T14:08:03.8408567Z warning_on_retry: true 2025-09-09T14:08:03.8408803Z continue_on_error: false 2025-09-09T14:08:03.8409050Z env: 2025-09-09T14:08:03.8409289Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:08:03.8409627Z REPOSITORY: pytorch/ao 2025-09-09T14:08:03.8409867Z PR_NUMBER: 2963 2025-09-09T14:08:03.8411250Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:08:03.8412715Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:08:03.8413260Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:08:03.8413762Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:08:03.8414193Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:08:03.8414619Z DRIVER_VERSION: 580.65.06 2025-09-09T14:08:03.8414871Z ##[endgroup] 2025-09-09T14:08:03.9756039Z == Installing nvidia driver NVIDIA-Linux-x86_64-580.65.06.run == 2025-09-09T14:08:03.9757615Z + pre_install_nvidia_driver_amzn2 2025-09-09T14:08:03.9761280Z + sudo yum remove -y nvidia-driver-latest-dkms 2025-09-09T14:08:04.3626791Z No match for argument: nvidia-driver-latest-dkms 2025-09-09T14:08:04.3630211Z No packages marked for removal. 2025-09-09T14:08:04.3684170Z Dependencies resolved. 2025-09-09T14:08:04.3694206Z Nothing to do. 2025-09-09T14:08:04.3694563Z Complete! 2025-09-09T14:08:04.5066536Z + install_nvidia_driver_common 2025-09-09T14:08:04.5070610Z + echo 'Before installing NVIDIA driver' 2025-09-09T14:08:04.5070901Z + lspci 2025-09-09T14:08:04.5072520Z Before installing NVIDIA driver 2025-09-09T14:08:04.5208746Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:08:04.5209244Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:08:04.5209807Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:08:04.5210317Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:08:04.5210790Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:08:04.5211299Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:08:04.5211761Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.5212176Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.5212579Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.5212985Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.5213444Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:08:04.5213828Z + lsmod 2025-09-09T14:08:04.5268102Z Module Size Used by 2025-09-09T14:08:04.5268415Z veth 36864 0 2025-09-09T14:08:04.5268671Z nvidia_modeset 1740800 0 2025-09-09T14:08:04.5268938Z video 65536 1 nvidia_modeset 2025-09-09T14:08:04.5269227Z wmi 36864 1 video 2025-09-09T14:08:04.5269497Z nvidia_uvm 1921024 0 2025-09-09T14:08:04.5269780Z nvidia 14274560 19 nvidia_uvm,nvidia_modeset 2025-09-09T14:08:04.5270094Z drm 602112 1 nvidia 2025-09-09T14:08:04.5270381Z drm_panel_orientation_quirks 32768 1 drm 2025-09-09T14:08:04.5270766Z backlight 24576 3 video,drm,nvidia_modeset 2025-09-09T14:08:04.5271214Z i2c_core 110592 2 nvidia,drm 2025-09-09T14:08:04.5271578Z xt_conntrack 16384 1 2025-09-09T14:08:04.5271894Z nft_chain_nat 16384 3 2025-09-09T14:08:04.5272211Z xt_MASQUERADE 20480 1 2025-09-09T14:08:04.5272549Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:08:04.5272867Z nf_conntrack_netlink 57344 0 2025-09-09T14:08:04.5273263Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:08:04.5273681Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:08:04.5273982Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:08:04.5274491Z xfrm_user 57344 1 2025-09-09T14:08:04.5274755Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:08:04.5275026Z xt_addrtype 16384 2 2025-09-09T14:08:04.5275275Z nft_compat 20480 4 2025-09-09T14:08:04.5275572Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:08:04.5275969Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:08:04.5276432Z br_netfilter 36864 0 2025-09-09T14:08:04.5276695Z bridge 323584 1 br_netfilter 2025-09-09T14:08:04.5276978Z stp 16384 1 bridge 2025-09-09T14:08:04.5277249Z llc 16384 2 bridge,stp 2025-09-09T14:08:04.5277526Z overlay 167936 0 2025-09-09T14:08:04.5277911Z tls 139264 0 2025-09-09T14:08:04.5278156Z nls_ascii 16384 1 2025-09-09T14:08:04.5278395Z nls_cp437 20480 1 2025-09-09T14:08:04.5278652Z vfat 24576 1 2025-09-09T14:08:04.5278903Z fat 86016 1 vfat 2025-09-09T14:08:04.5279154Z sunrpc 700416 1 2025-09-09T14:08:04.5279395Z i8042 45056 0 2025-09-09T14:08:04.5279627Z ena 180224 0 2025-09-09T14:08:04.5279875Z serio 28672 3 i8042 2025-09-09T14:08:04.5280140Z ghash_clmulni_intel 16384 0 2025-09-09T14:08:04.5280391Z button 24576 0 2025-09-09T14:08:04.5280637Z sch_fq_codel 20480 33 2025-09-09T14:08:04.5280880Z fuse 184320 1 2025-09-09T14:08:04.5281116Z dm_mod 188416 0 2025-09-09T14:08:04.5281348Z loop 36864 0 2025-09-09T14:08:04.5281590Z configfs 57344 1 2025-09-09T14:08:04.5281834Z dmi_sysfs 20480 0 2025-09-09T14:08:04.5282081Z crc32_pclmul 16384 0 2025-09-09T14:08:04.5282321Z crc32c_intel 24576 0 2025-09-09T14:08:04.5282564Z efivarfs 24576 1 2025-09-09T14:08:04.5282799Z + modinfo nvidia 2025-09-09T14:08:04.5292950Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:08:04.5293394Z import_ns: DMA_BUF 2025-09-09T14:08:04.5293631Z alias: char-major-195-* 2025-09-09T14:08:04.5293973Z version: 580.65.06 2025-09-09T14:08:04.5294253Z supported: external 2025-09-09T14:08:04.5294575Z license: Dual MIT/GPL 2025-09-09T14:08:04.5294855Z firmware: nvidia/580.65.06/gsp_tu10x.bin 2025-09-09T14:08:04.5295176Z firmware: nvidia/580.65.06/gsp_ga10x.bin 2025-09-09T14:08:04.5295489Z srcversion: A69EBF72FC9D60E11E9A05C 2025-09-09T14:08:04.5295803Z alias: of:N*T*Cnvidia,tegra264-displayC* 2025-09-09T14:08:04.5296144Z alias: of:N*T*Cnvidia,tegra264-display 2025-09-09T14:08:04.5296485Z alias: of:N*T*Cnvidia,tegra234-displayC* 2025-09-09T14:08:04.5296821Z alias: of:N*T*Cnvidia,tegra234-display 2025-09-09T14:08:04.5297146Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:08:04.5297479Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:08:04.5297803Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:08:04.5298099Z depends: i2c-core,drm 2025-09-09T14:08:04.5298346Z retpoline: Y 2025-09-09T14:08:04.5298547Z name: nvidia 2025-09-09T14:08:04.5298899Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:08:04.5299353Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:08:04.5299791Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:08:04.5300194Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:08:04.5300497Z parm: NVreg_RmLogonRC:int 2025-09-09T14:08:04.5300797Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:08:04.5301098Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:08:04.5301393Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:08:04.5301683Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:08:04.5302153Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:08:04.5302531Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:08:04.5302858Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:08:04.5303145Z parm: NVreg_EnableMSI:int 2025-09-09T14:08:04.5303443Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:08:04.5303796Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:08:04.5304180Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:08:04.5304553Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:08:04.5304951Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:04.5305429Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:08:04.5305836Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:04.5306240Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:08:04.5306566Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:08:04.5306924Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:08:04.5307290Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:08:04.5307615Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:08:04.5307931Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:08:04.5308250Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:08:04.5308614Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:08:04.5308909Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:08:04.5309246Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:08:04.5309590Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:08:04.5309937Z parm: NVreg_RegisterPlatformDeviceDriver:int 2025-09-09T14:08:04.5310293Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:08:04.5310612Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:08:04.5310948Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:08:04.5311290Z parm: NVreg_CoherentGPUMemoryMode:charp 2025-09-09T14:08:04.5311626Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:08:04.5311950Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:08:04.5312276Z parm: NVreg_RmMsg:charp 2025-09-09T14:08:04.5312551Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:08:04.5312869Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:08:04.5313183Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:08:04.5313486Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:08:04.5313809Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:08:04.5314151Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:08:04.5314504Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:08:04.5314815Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:08:04.5315151Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:08:04.5315478Z parm: rm_firmware_active:charp 2025-09-09T14:08:04.5315783Z + HAS_NVIDIA_DRIVER=0 2025-09-09T14:08:04.5316018Z ++ command -v nvidia-smi 2025-09-09T14:08:04.5316351Z + '[' -x /usr/bin/nvidia-smi ']' 2025-09-09T14:08:04.5316599Z + set +e 2025-09-09T14:08:04.5316895Z ++ nvidia-smi --query-gpu=driver_version --format=csv,noheader --id=0 2025-09-09T14:08:04.5910920Z + INSTALLED_DRIVER_VERSION=580.65.06 2025-09-09T14:08:04.5911213Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:08:04.5911533Z + '[' 0 -ne 0 ']' 2025-09-09T14:08:04.5911749Z + '[' 580.65.06 '!=' 580.65.06 ']' 2025-09-09T14:08:04.5911998Z + HAS_NVIDIA_DRIVER=1 2025-09-09T14:08:04.5912419Z + echo 'NVIDIA driver (580.65.06) has already been installed. Skipping NVIDIA driver installation' 2025-09-09T14:08:04.5912874Z + set -e 2025-09-09T14:08:04.5913064Z + '[' 1 -eq 0 ']' 2025-09-09T14:08:04.5913428Z NVIDIA driver (580.65.06) has already been installed. Skipping NVIDIA driver installation 2025-09-09T14:08:04.5914174Z + post_install_nvidia_driver_common 2025-09-09T14:08:04.5920113Z + sudo modprobe nvidia 2025-09-09T14:08:04.7271408Z + echo 'After installing NVIDIA driver' 2025-09-09T14:08:04.7271721Z + lspci 2025-09-09T14:08:04.7271930Z After installing NVIDIA driver 2025-09-09T14:08:04.7402753Z 00:00.0 Host bridge: Intel Corporation 440FX - 82441FX PMC [Natoma] 2025-09-09T14:08:04.7403242Z 00:01.0 ISA bridge: Intel Corporation 82371SB PIIX3 ISA [Natoma/Triton II] 2025-09-09T14:08:04.7403782Z 00:01.3 Non-VGA unclassified device: Intel Corporation 82371AB/EB/MB PIIX4 ACPI (rev 08) 2025-09-09T14:08:04.7404291Z 00:03.0 VGA compatible controller: Amazon.com, Inc. Device 1111 2025-09-09T14:08:04.7404754Z 00:04.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe EBS Controller 2025-09-09T14:08:04.7405273Z 00:05.0 Ethernet controller: Amazon.com, Inc. Elastic Network Adapter (ENA) 2025-09-09T14:08:04.7405907Z 00:1b.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.7406323Z 00:1c.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.7406747Z 00:1d.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.7407150Z 00:1e.0 3D controller: NVIDIA Corporation GA102GL [A10G] (rev a1) 2025-09-09T14:08:04.7407617Z 00:1f.0 Non-Volatile memory controller: Amazon.com, Inc. NVMe SSD Controller 2025-09-09T14:08:04.7408010Z + lsmod 2025-09-09T14:08:04.7445282Z Module Size Used by 2025-09-09T14:08:04.7445854Z veth 36864 0 2025-09-09T14:08:04.7446332Z nvidia_modeset 1740800 0 2025-09-09T14:08:04.7446861Z video 65536 1 nvidia_modeset 2025-09-09T14:08:04.7447423Z wmi 36864 1 video 2025-09-09T14:08:04.7447941Z nvidia_uvm 1921024 0 2025-09-09T14:08:04.7448509Z nvidia 14274560 19 nvidia_uvm,nvidia_modeset 2025-09-09T14:08:04.7448886Z drm 602112 1 nvidia 2025-09-09T14:08:04.7449204Z drm_panel_orientation_quirks 32768 1 drm 2025-09-09T14:08:04.7449544Z backlight 24576 3 video,drm,nvidia_modeset 2025-09-09T14:08:04.7449884Z i2c_core 110592 2 nvidia,drm 2025-09-09T14:08:04.7450156Z xt_conntrack 16384 1 2025-09-09T14:08:04.7450410Z nft_chain_nat 16384 3 2025-09-09T14:08:04.7450652Z xt_MASQUERADE 20480 1 2025-09-09T14:08:04.7450939Z nf_nat 57344 2 nft_chain_nat,xt_MASQUERADE 2025-09-09T14:08:04.7451279Z nf_conntrack_netlink 57344 0 2025-09-09T14:08:04.7451662Z nf_conntrack 184320 4 xt_conntrack,nf_nat,nf_conntrack_netlink,xt_MASQUERADE 2025-09-09T14:08:04.7452078Z nf_defrag_ipv6 24576 1 nf_conntrack 2025-09-09T14:08:04.7452376Z nf_defrag_ipv4 16384 1 nf_conntrack 2025-09-09T14:08:04.7452649Z xfrm_user 57344 1 2025-09-09T14:08:04.7452916Z xfrm_algo 16384 1 xfrm_user 2025-09-09T14:08:04.7453191Z xt_addrtype 16384 2 2025-09-09T14:08:04.7453438Z nft_compat 20480 4 2025-09-09T14:08:04.7453724Z nf_tables 311296 57 nft_compat,nft_chain_nat 2025-09-09T14:08:04.7454128Z nfnetlink 20480 4 nft_compat,nf_conntrack_netlink,nf_tables 2025-09-09T14:08:04.7454494Z br_netfilter 36864 0 2025-09-09T14:08:04.7454754Z bridge 323584 1 br_netfilter 2025-09-09T14:08:04.7455041Z stp 16384 1 bridge 2025-09-09T14:08:04.7455307Z llc 16384 2 bridge,stp 2025-09-09T14:08:04.7455582Z overlay 167936 0 2025-09-09T14:08:04.7467792Z tls 139264 0 2025-09-09T14:08:04.7468086Z nls_ascii 16384 1 2025-09-09T14:08:04.7468334Z nls_cp437 20480 1 2025-09-09T14:08:04.7468573Z vfat 24576 1 2025-09-09T14:08:04.7468802Z fat 86016 1 vfat 2025-09-09T14:08:04.7469061Z sunrpc 700416 1 2025-09-09T14:08:04.7469290Z i8042 45056 0 2025-09-09T14:08:04.7469527Z ena 180224 0 2025-09-09T14:08:04.7469765Z serio 28672 3 i8042 2025-09-09T14:08:04.7470017Z ghash_clmulni_intel 16384 0 2025-09-09T14:08:04.7470429Z button 24576 0 2025-09-09T14:08:04.7470668Z sch_fq_codel 20480 33 2025-09-09T14:08:04.7470904Z fuse 184320 1 2025-09-09T14:08:04.7471131Z dm_mod 188416 0 2025-09-09T14:08:04.7471363Z loop 36864 0 2025-09-09T14:08:04.7471603Z configfs 57344 1 2025-09-09T14:08:04.7471855Z dmi_sysfs 20480 0 2025-09-09T14:08:04.7472096Z crc32_pclmul 16384 0 2025-09-09T14:08:04.7472345Z crc32c_intel 24576 0 2025-09-09T14:08:04.7472594Z efivarfs 24576 1 2025-09-09T14:08:04.7472860Z + modinfo nvidia 2025-09-09T14:08:04.7473237Z filename: /lib/modules/6.1.141-155.222.amzn2023.x86_64/kernel/drivers/video/nvidia.ko 2025-09-09T14:08:04.7473787Z import_ns: DMA_BUF 2025-09-09T14:08:04.7474025Z alias: char-major-195-* 2025-09-09T14:08:04.7474285Z version: 580.65.06 2025-09-09T14:08:04.7474516Z supported: external 2025-09-09T14:08:04.7474768Z license: Dual MIT/GPL 2025-09-09T14:08:04.7475041Z firmware: nvidia/580.65.06/gsp_tu10x.bin 2025-09-09T14:08:04.7475371Z firmware: nvidia/580.65.06/gsp_ga10x.bin 2025-09-09T14:08:04.7475681Z srcversion: A69EBF72FC9D60E11E9A05C 2025-09-09T14:08:04.7476007Z alias: of:N*T*Cnvidia,tegra264-displayC* 2025-09-09T14:08:04.7476434Z alias: of:N*T*Cnvidia,tegra264-display 2025-09-09T14:08:04.7476775Z alias: of:N*T*Cnvidia,tegra234-displayC* 2025-09-09T14:08:04.7477114Z alias: of:N*T*Cnvidia,tegra234-display 2025-09-09T14:08:04.7477437Z alias: pci:v000010DEd*sv*sd*bc06sc80i00* 2025-09-09T14:08:04.7477766Z alias: pci:v000010DEd*sv*sd*bc03sc02i00* 2025-09-09T14:08:04.7478089Z alias: pci:v000010DEd*sv*sd*bc03sc00i00* 2025-09-09T14:08:04.7478381Z depends: i2c-core,drm 2025-09-09T14:08:04.7478622Z retpoline: Y 2025-09-09T14:08:04.7478833Z name: nvidia 2025-09-09T14:08:04.7479180Z vermagic: 6.1.141-155.222.amzn2023.x86_64 SMP preempt mod_unload modversions 2025-09-09T14:08:04.7479641Z parm: NvSwitchRegDwords:NvSwitch regkey (charp) 2025-09-09T14:08:04.7480071Z parm: NvSwitchBlacklist:NvSwitchBlacklist=uuid[,uuid...] (charp) 2025-09-09T14:08:04.7480471Z parm: NVreg_ResmanDebugLevel:int 2025-09-09T14:08:04.7480763Z parm: NVreg_RmLogonRC:int 2025-09-09T14:08:04.7481047Z parm: NVreg_ModifyDeviceFiles:int 2025-09-09T14:08:04.7481350Z parm: NVreg_DeviceFileUID:int 2025-09-09T14:08:04.7481632Z parm: NVreg_DeviceFileGID:int 2025-09-09T14:08:04.7481931Z parm: NVreg_DeviceFileMode:int 2025-09-09T14:08:04.7482276Z parm: NVreg_InitializeSystemMemoryAllocations:int 2025-09-09T14:08:04.7482656Z parm: NVreg_UsePageAttributeTable:int 2025-09-09T14:08:04.7482975Z parm: NVreg_EnablePCIeGen3:int 2025-09-09T14:08:04.7483257Z parm: NVreg_EnableMSI:int 2025-09-09T14:08:04.7483558Z parm: NVreg_EnableStreamMemOPs:int 2025-09-09T14:08:04.7483905Z parm: NVreg_RestrictProfilingToAdminUsers:int 2025-09-09T14:08:04.7484292Z parm: NVreg_PreserveVideoMemoryAllocations:int 2025-09-09T14:08:04.7484656Z parm: NVreg_EnableS0ixPowerManagement:int 2025-09-09T14:08:04.7485058Z parm: NVreg_S0ixPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:04.7485447Z parm: NVreg_DynamicPowerManagement:int 2025-09-09T14:08:04.7485857Z parm: NVreg_DynamicPowerManagementVideoMemoryThreshold:int 2025-09-09T14:08:04.7486254Z parm: NVreg_EnableGpuFirmware:int 2025-09-09T14:08:04.7486575Z parm: NVreg_EnableGpuFirmwareLogs:int 2025-09-09T14:08:04.7486935Z parm: NVreg_OpenRmEnableUnsupportedGpus:int 2025-09-09T14:08:04.7487294Z parm: NVreg_EnableUserNUMAManagement:int 2025-09-09T14:08:04.7487626Z parm: NVreg_MemoryPoolSize:int 2025-09-09T14:08:04.7487930Z parm: NVreg_KMallocHeapMaxSize:int 2025-09-09T14:08:04.7488351Z parm: NVreg_VMallocHeapMaxSize:int 2025-09-09T14:08:04.7488679Z parm: NVreg_IgnoreMMIOCheck:int 2025-09-09T14:08:04.7488979Z parm: NVreg_NvLinkDisable:int 2025-09-09T14:08:04.7489318Z parm: NVreg_EnablePCIERelaxedOrderingMode:int 2025-09-09T14:08:04.7489663Z parm: NVreg_RegisterPCIDriver:int 2025-09-09T14:08:04.7490013Z parm: NVreg_RegisterPlatformDeviceDriver:int 2025-09-09T14:08:04.7490357Z parm: NVreg_EnableResizableBar:int 2025-09-09T14:08:04.7490681Z parm: NVreg_EnableDbgBreakpoint:int 2025-09-09T14:08:04.7491011Z parm: NVreg_EnableNonblockingOpen:int 2025-09-09T14:08:04.7491352Z parm: NVreg_CoherentGPUMemoryMode:charp 2025-09-09T14:08:04.7491762Z parm: NVreg_RegistryDwords:charp 2025-09-09T14:08:04.7492083Z parm: NVreg_RegistryDwordsPerDevice:charp 2025-09-09T14:08:04.7492406Z parm: NVreg_RmMsg:charp 2025-09-09T14:08:04.7492676Z parm: NVreg_GpuBlacklist:charp 2025-09-09T14:08:04.7492998Z parm: NVreg_TemporaryFilePath:charp 2025-09-09T14:08:04.7493306Z parm: NVreg_ExcludedGpus:charp 2025-09-09T14:08:04.7493611Z parm: NVreg_DmaRemapPeerMmio:int 2025-09-09T14:08:04.7493922Z parm: NVreg_RmNvlinkBandwidth:charp 2025-09-09T14:08:04.7494267Z parm: NVreg_RmNvlinkBandwidthLinkCount:int 2025-09-09T14:08:04.7494607Z parm: NVreg_ImexChannelCount:int 2025-09-09T14:08:04.7494916Z parm: NVreg_CreateImexChannel0:int 2025-09-09T14:08:04.7495254Z parm: NVreg_GrdmaPciTopoCheckOverride:int 2025-09-09T14:08:04.7495574Z parm: rm_firmware_active:charp 2025-09-09T14:08:04.7495851Z + set +e 2025-09-09T14:08:04.7496030Z + nvidia-smi 2025-09-09T14:08:04.7910523Z Tue Sep 9 14:08:04 2025 2025-09-09T14:08:04.7911190Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:04.7912142Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:08:04.7913057Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:04.7913981Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:08:04.7914988Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:08:04.7915796Z | | | MIG M. | 2025-09-09T14:08:04.7916569Z |=========================================+========================+======================| 2025-09-09T14:08:04.8508803Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:08:04.8509261Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:04.8509639Z | | | N/A | 2025-09-09T14:08:04.8510021Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:04.8510465Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:08:04.8510880Z | 0% 24C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:04.8511238Z | | | N/A | 2025-09-09T14:08:04.8511627Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:04.8512060Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:08:04.8512490Z | 0% 23C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:04.8512866Z | | | N/A | 2025-09-09T14:08:04.8513439Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:04.8513878Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:08:04.8514286Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:04.8514654Z | | | N/A | 2025-09-09T14:08:04.8515035Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:04.8535412Z 2025-09-09T14:08:04.8535717Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:04.8536147Z | Processes: | 2025-09-09T14:08:04.8536718Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:08:04.8537136Z | ID ID Usage | 2025-09-09T14:08:04.8537485Z |=========================================================================================| 2025-09-09T14:08:04.8574213Z | No running processes found | 2025-09-09T14:08:04.8574683Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:05.9302256Z + nvidia-smi --query-gpu=gpu_name --format=csv,noheader --id=0 2025-09-09T14:08:05.9483162Z NVIDIA A10G 2025-09-09T14:08:05.9766371Z + NVIDIA_SMI_STATUS=0 2025-09-09T14:08:05.9766606Z + '[' 0 -eq 0 ']' 2025-09-09T14:08:05.9766849Z + echo 'INFO: Ignoring allowed status 0' 2025-09-09T14:08:05.9767122Z + set -e 2025-09-09T14:08:05.9767333Z INFO: Ignoring allowed status 0 2025-09-09T14:08:05.9777103Z == Installing nvidia container toolkit for amzn2023 == 2025-09-09T14:08:05.9781682Z + sudo yum install -y yum-utils 2025-09-09T14:08:06.4593104Z Last metadata expiration check: 0:03:40 ago on Tue Sep 9 14:04:26 2025. 2025-09-09T14:08:06.4843983Z Package dnf-utils-4.3.0-13.amzn2023.0.5.noarch is already installed. 2025-09-09T14:08:06.5306162Z Dependencies resolved. 2025-09-09T14:08:06.5534568Z Nothing to do. 2025-09-09T14:08:06.5534880Z Complete! 2025-09-09T14:08:06.6893854Z + [[ amzn2023 == \a\m\z\n\2\0\2\3 ]] 2025-09-09T14:08:06.6894410Z + YUM_REPO_URL=https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:08:06.6895255Z + sudo yum-config-manager --add-repo https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:08:06.9489475Z Adding repo from: https://nvidia.github.io/libnvidia-container/stable/rpm/nvidia-container-toolkit.repo 2025-09-09T14:08:06.9988219Z + sudo yum install -y nvidia-container-toolkit-1.17.8 libnvidia-container-tools-1.17.8 libnvidia-container1-1.17.8 nvidia-container-toolkit-base-1.17.8 2025-09-09T14:08:07.5182252Z nvidia-container-toolkit 18 kB/s | 833 B 00:00 2025-09-09T14:08:07.5428614Z Package nvidia-container-toolkit-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:07.5434061Z Package libnvidia-container-tools-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:07.5437896Z Package libnvidia-container1-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:07.5444518Z Package nvidia-container-toolkit-base-1.17.8-1.x86_64 is already installed. 2025-09-09T14:08:07.5916633Z Dependencies resolved. 2025-09-09T14:08:07.6139713Z Nothing to do. 2025-09-09T14:08:07.6140057Z Complete! 2025-09-09T14:08:07.7515967Z + sudo systemctl restart docker 2025-09-09T14:08:46.9042424Z nvidia-persistenced failed to initialize. Check syslog for more details. 2025-09-09T14:08:46.9506992Z Tue Sep 9 14:08:46 2025 2025-09-09T14:08:46.9507388Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:46.9507880Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:08:46.9508739Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:46.9509233Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:08:46.9509751Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:08:46.9510178Z | | | MIG M. | 2025-09-09T14:08:46.9510496Z |=========================================+========================+======================| 2025-09-09T14:08:47.0108820Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:08:47.0109277Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:47.0109858Z | | | N/A | 2025-09-09T14:08:47.0110249Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:47.0110685Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:08:47.0111090Z | 0% 24C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:47.0111456Z | | | N/A | 2025-09-09T14:08:47.0111835Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:47.0112266Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:08:47.0112670Z | 0% 23C P8 11W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:47.0113041Z | | | N/A | 2025-09-09T14:08:47.0113421Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:47.0113850Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:08:47.0114267Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:08:47.0114624Z | | | N/A | 2025-09-09T14:08:47.0115017Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:08:47.0135706Z 2025-09-09T14:08:47.0135928Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:47.0136356Z | Processes: | 2025-09-09T14:08:47.0136793Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:08:47.0137187Z | ID ID Usage | 2025-09-09T14:08:47.0137530Z |=========================================================================================| 2025-09-09T14:08:47.0172394Z | No running processes found | 2025-09-09T14:08:47.0172847Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:08:48.0856908Z Unable to find image 'public.ecr.aws/docker/library/python:3.13' locally 2025-09-09T14:08:48.3349056Z 3.13: Pulling from docker/library/python 2025-09-09T14:08:48.4291708Z 15b1d8a5ff03: Pulling fs layer 2025-09-09T14:08:48.4291986Z 22718812f617: Pulling fs layer 2025-09-09T14:08:48.4292494Z 401a98f7495b: Pulling fs layer 2025-09-09T14:08:48.4292849Z ad446e7df19a: Pulling fs layer 2025-09-09T14:08:48.4293202Z 5d32990caa16: Pulling fs layer 2025-09-09T14:08:48.4293537Z a79d633abf9a: Pulling fs layer 2025-09-09T14:08:48.4293848Z 249a56c8e466: Pulling fs layer 2025-09-09T14:08:48.4294146Z 249a56c8e466: Waiting 2025-09-09T14:08:48.4294409Z a79d633abf9a: Waiting 2025-09-09T14:08:48.4296186Z ad446e7df19a: Waiting 2025-09-09T14:08:48.4296492Z 5d32990caa16: Waiting 2025-09-09T14:08:48.5681175Z 22718812f617: Verifying Checksum 2025-09-09T14:08:48.5681520Z 22718812f617: Download complete 2025-09-09T14:08:48.6267201Z 15b1d8a5ff03: Verifying Checksum 2025-09-09T14:08:48.6267796Z 15b1d8a5ff03: Download complete 2025-09-09T14:08:48.7086773Z 5d32990caa16: Verifying Checksum 2025-09-09T14:08:48.7087071Z 5d32990caa16: Download complete 2025-09-09T14:08:48.7668424Z 401a98f7495b: Verifying Checksum 2025-09-09T14:08:48.7668711Z 401a98f7495b: Download complete 2025-09-09T14:08:48.8338949Z a79d633abf9a: Download complete 2025-09-09T14:08:48.8467656Z 249a56c8e466: Download complete 2025-09-09T14:08:49.3178326Z ad446e7df19a: Verifying Checksum 2025-09-09T14:08:49.3178838Z ad446e7df19a: Download complete 2025-09-09T14:08:50.3658682Z 15b1d8a5ff03: Pull complete 2025-09-09T14:08:51.0986610Z 22718812f617: Pull complete 2025-09-09T14:08:53.5449402Z 401a98f7495b: Pull complete 2025-09-09T14:09:00.3690167Z ad446e7df19a: Pull complete 2025-09-09T14:09:00.6599925Z 5d32990caa16: Pull complete 2025-09-09T14:09:01.4326507Z a79d633abf9a: Pull complete 2025-09-09T14:09:01.4555181Z 249a56c8e466: Pull complete 2025-09-09T14:09:01.4690161Z Digest: sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T14:09:01.4733046Z Status: Downloaded newer image for public.ecr.aws/docker/library/python:3.13 2025-09-09T14:09:08.1891833Z Tue Sep 9 14:09:08 2025 2025-09-09T14:09:08.1892336Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:09:08.1892870Z | NVIDIA-SMI 580.65.06 Driver Version: 580.65.06 CUDA Version: 13.0 | 2025-09-09T14:09:08.1893340Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:08.1893840Z | GPU Name Persistence-M | Bus-Id Disp.A | Volatile Uncorr. ECC | 2025-09-09T14:09:08.1894368Z | Fan Temp Perf Pwr:Usage/Cap | Memory-Usage | GPU-Util Compute M. | 2025-09-09T14:09:08.1894780Z | | | MIG M. | 2025-09-09T14:09:08.1895103Z |=========================================+========================+======================| 2025-09-09T14:09:08.2503083Z | 0 NVIDIA A10G On | 00000000:00:1B.0 Off | 0 | 2025-09-09T14:09:08.2504232Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:08.2505160Z | | | N/A | 2025-09-09T14:09:08.2505898Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:08.2506772Z | 1 NVIDIA A10G On | 00000000:00:1C.0 Off | 0 | 2025-09-09T14:09:08.2507710Z | 0% 24C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:08.2508419Z | | | N/A | 2025-09-09T14:09:08.2509038Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:08.2509515Z | 2 NVIDIA A10G On | 00000000:00:1D.0 Off | 0 | 2025-09-09T14:09:08.2509935Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:08.2510294Z | | | N/A | 2025-09-09T14:09:08.2510683Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:08.2511118Z | 3 NVIDIA A10G On | 00000000:00:1E.0 Off | 0 | 2025-09-09T14:09:08.2511535Z | 0% 23C P8 10W / 300W | 0MiB / 23028MiB | 0% Default | 2025-09-09T14:09:08.2512249Z | | | N/A | 2025-09-09T14:09:08.2512638Z +-----------------------------------------+------------------------+----------------------+ 2025-09-09T14:09:08.2528555Z 2025-09-09T14:09:08.2528837Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:09:08.2529410Z | Processes: | 2025-09-09T14:09:08.2529905Z | GPU GI CI PID Type Process name GPU Memory | 2025-09-09T14:09:08.2530304Z | ID ID Usage | 2025-09-09T14:09:08.2530626Z |=========================================================================================| 2025-09-09T14:09:08.2566141Z | No running processes found | 2025-09-09T14:09:08.2566785Z +-----------------------------------------------------------------------------------------+ 2025-09-09T14:09:09.9524237Z Command completed after 1 attempt(s). 2025-09-09T14:09:09.9632302Z ##[group]Run set -ex 2025-09-09T14:09:09.9632577Z set -ex 2025-09-09T14:09:09.9632770Z { 2025-09-09T14:09:09.9632985Z  echo "#!/usr/bin/env bash"; 2025-09-09T14:09:09.9633278Z  echo "set -eou pipefail"; 2025-09-09T14:09:09.9633557Z  # shellcheck disable=SC2016 2025-09-09T14:09:09.9633867Z  echo 'eval "$(conda shell.bash hook)"'; 2025-09-09T14:09:09.9634163Z  echo "set -x"; 2025-09-09T14:09:09.9634403Z  echo "${SCRIPT}"; 2025-09-09T14:09:09.9634657Z } > "${RUNNER_TEMP}/exec_script" 2025-09-09T14:09:09.9634970Z chmod +x "${RUNNER_TEMP}/exec_script" 2025-09-09T14:09:09.9635563Z python3 "/home/ec2-user/actions-runner/_work/ao/ao/test-infra/.github/scripts/run_with_env_secrets.py" "" 2025-09-09T14:09:09.9650922Z shell: /usr/bin/bash -e {0} 2025-09-09T14:09:09.9651163Z env: 2025-09-09T14:09:09.9651402Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T14:09:09.9651747Z REPOSITORY: pytorch/ao 2025-09-09T14:09:09.9651984Z PR_NUMBER: 2963 2025-09-09T14:09:09.9653282Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T14:09:09.9654737Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T14:09:09.9655263Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T14:09:09.9655774Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T14:09:09.9667237Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T14:09:09.9667833Z ALL_SECRETS: { "github_token": "***" } 2025-09-09T14:09:09.9668122Z ##[endgroup] 2025-09-09T14:09:09.9709174Z + echo '#!/usr/bin/env bash' 2025-09-09T14:09:09.9709459Z + echo 'set -eou pipefail' 2025-09-09T14:09:09.9709718Z + echo 'eval "$(conda shell.bash hook)"' 2025-09-09T14:09:09.9709999Z + echo 'set -x' 2025-09-09T14:09:09.9710229Z + echo 'conda create -n venv python=3.9 -y 2025-09-09T14:09:09.9710518Z conda activate venv 2025-09-09T14:09:09.9710796Z python -m pip install --upgrade pip 2025-09-09T14:09:09.9711237Z pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 2025-09-09T14:09:09.9711688Z pip install -r dev-requirements.txt 2025-09-09T14:09:09.9711955Z pip install . 2025-09-09T14:09:09.9712193Z export CONDA=$(dirname $(dirname $(which conda))) 2025-09-09T14:09:09.9712560Z export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH 2025-09-09T14:09:09.9712879Z pytest test --verbose -s 2025-09-09T14:09:09.9713105Z ' 2025-09-09T14:09:09.9713378Z + chmod +x /home/ec2-user/actions-runner/_work/_temp/exec_script 2025-09-09T14:09:09.9727634Z + python3 /home/ec2-user/actions-runner/_work/ao/ao/test-infra/.github/scripts/run_with_env_secrets.py '' 2025-09-09T14:09:17.3255984Z Running command: 2025-09-09T14:09:17.3263508Z docker run -e PR_NUMBER -e RUNNER_ARTIFACT_DIR=/artifacts -e RUNNER_DOCS_DIR=/docs -e RUNNER_TEST_RESULTS_DIR=/test-results --env-file="/home/ec2-user/actions-runner/_work/_temp/github_env_17585175130" `# It is unknown why the container sees a different value for this.` -e GITHUB_STEP_SUMMARY -e SECRET_GITHUB_TOKEN --cap-add=SYS_PTRACE --detach --ipc=host --security-opt seccomp=unconfined --shm-size=2g --tty --ulimit stack=10485760:83886080 --ulimit core=0 --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all -v "/home/ec2-user/actions-runner/_work/ao/ao/pytorch/ao:/pytorch/ao" -v "/home/ec2-user/actions-runner/_work/ao/ao/test-infra:/test-infra" -v "/home/ec2-user/actions-runner/_work/_temp/artifacts:/artifacts" -v "/home/ec2-user/actions-runner/_work/_temp/docs:/docs" -v "/home/ec2-user/actions-runner/_work/_temp/test-results:/test-results" -v "/home/ec2-user/actions-runner/_work/_temp/exec_script:/exec" -v "/home/ec2-user/actions-runner/_work/_temp/_runner_file_commands/step_summary_3221eec8-936f-42e3-a0c4-9c210cb864fa":"/home/ec2-user/actions-runner/_work/_temp/_runner_file_commands/step_summary_3221eec8-936f-42e3-a0c4-9c210cb864fa" -w /pytorch/ao "pytorch/almalinux-builder:cuda12.6" 2025-09-09T14:09:17.3269900Z 2025-09-09T14:09:17.3270277Z 6f14d54195719389a6ca04d74d944db4d0b5bcd49ba6fe5591c3d08f98564f8f 2025-09-09T14:09:17.3271112Z Running command: docker exec -t 6f14d54195719389a6ca04d74d944db4d0b5bcd49ba6fe5591c3d08f98564f8f /exec 2025-09-09T14:09:17.3271758Z + conda create -n venv python=3.9 -y 2025-09-09T14:09:17.3272117Z + local cmd=create 2025-09-09T14:09:17.3272360Z + case "$cmd" in 2025-09-09T14:09:17.3272678Z + __conda_exe create -n venv python=3.9 -y 2025-09-09T14:09:17.3273164Z + /opt/conda/bin/conda create -n venv python=3.9 -y 2025-09-09T14:09:17.3274349Z Collecting package metadata (current_repodata.json): - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / - \ | / done 2025-09-09T14:09:17.3275083Z Solving environment: \ done 2025-09-09T14:09:17.3275316Z 2025-09-09T14:09:17.3275320Z 2025-09-09T14:09:17.3275494Z ==> WARNING: A newer version of conda exists. <== 2025-09-09T14:09:17.3275884Z current version: 23.5.2 2025-09-09T14:09:17.3276135Z latest version: 25.7.0 2025-09-09T14:09:17.3276444Z 2025-09-09T14:09:17.3276551Z Please update conda by running 2025-09-09T14:09:17.3276720Z 2025-09-09T14:09:17.3276854Z $ conda update -n base -c defaults conda 2025-09-09T14:09:17.3277113Z 2025-09-09T14:09:17.3277320Z Or to minimize the number of packages updated during conda update use 2025-09-09T14:09:17.3277656Z 2025-09-09T14:09:17.3277789Z conda install conda=25.7.0 2025-09-09T14:09:17.3278033Z 2025-09-09T14:09:17.3278039Z 2025-09-09T14:09:17.3278045Z 2025-09-09T14:09:17.3278165Z ## Package Plan ## 2025-09-09T14:09:17.3278318Z 2025-09-09T14:09:17.3278485Z environment location: /opt/conda/envs/venv 2025-09-09T14:09:17.3278781Z 2025-09-09T14:09:17.3278891Z added / updated specs: 2025-09-09T14:09:17.3279160Z - python=3.9 2025-09-09T14:09:17.3279335Z 2025-09-09T14:09:17.3279339Z 2025-09-09T14:09:17.3279455Z The following packages will be downloaded: 2025-09-09T14:09:17.3279680Z 2025-09-09T14:09:17.3279788Z package | build 2025-09-09T14:09:17.3280089Z ---------------------------|----------------- 2025-09-09T14:09:17.3280451Z bzip2-1.0.8 | h5eee18b_6 262 KB 2025-09-09T14:09:17.3280954Z ld_impl_linux-64-2.40 | h12ee557_0 710 KB 2025-09-09T14:09:17.3281396Z libffi-3.4.4 | h6a678d5_1 141 KB 2025-09-09T14:09:17.3281832Z libxcb-1.17.0 | h9b100fa_0 430 KB 2025-09-09T14:09:17.3282288Z ncurses-6.5 | h7934f7d_0 1.1 MB 2025-09-09T14:09:17.3282794Z pip-25.2 | pyhc872135_0 1.2 MB 2025-09-09T14:09:17.3283280Z pthread-stubs-0.3 | h0ce48e5_1 5 KB 2025-09-09T14:09:17.3283814Z python-3.9.23 | he99959a_0 24.7 MB 2025-09-09T14:09:17.3284203Z readline-8.3 | hc2a1206_0 471 KB 2025-09-09T14:09:17.3284998Z setuptools-78.1.1 | py39h06a4308_0 1.7 MB 2025-09-09T14:09:17.3285481Z sqlite-3.50.2 | hb25bd0a_1 1.1 MB 2025-09-09T14:09:17.3285959Z tk-8.6.15 | h54e0aa7_0 3.4 MB 2025-09-09T14:09:17.3286473Z tzdata-2025b | h04d1e81_0 116 KB 2025-09-09T14:09:17.3287159Z wheel-0.45.1 | py39h06a4308_0 114 KB 2025-09-09T14:09:17.3287610Z xorg-libx11-1.8.12 | h9b100fa_1 895 KB 2025-09-09T14:09:17.3288077Z xorg-libxau-1.0.12 | h9b100fa_0 13 KB 2025-09-09T14:09:17.3288533Z xorg-libxdmcp-1.1.5 | h9b100fa_0 19 KB 2025-09-09T14:09:17.3289098Z xorg-xorgproto-2024.1 | h5eee18b_1 580 KB 2025-09-09T14:09:17.3289645Z xz-5.6.4 | h5eee18b_1 567 KB 2025-09-09T14:09:17.3290169Z zlib-1.2.13 | h5eee18b_1 111 KB 2025-09-09T14:09:17.3290690Z ------------------------------------------------------------ 2025-09-09T14:09:17.3291100Z Total: 37.6 MB 2025-09-09T14:09:17.3291376Z 2025-09-09T14:09:17.3291545Z The following NEW packages will be INSTALLED: 2025-09-09T14:09:17.3291847Z 2025-09-09T14:09:17.3292117Z _libgcc_mutex pkgs/main/linux-64::_libgcc_mutex-0.1-main 2025-09-09T14:09:17.3292554Z _openmp_mutex pkgs/main/linux-64::_openmp_mutex-5.1-1_gnu 2025-09-09T14:09:17.3292957Z bzip2 pkgs/main/linux-64::bzip2-1.0.8-h5eee18b_6 2025-09-09T14:09:17.3293542Z ca-certificates pkgs/main/linux-64::ca-certificates-2025.7.15-h06a4308_0 2025-09-09T14:09:17.3294092Z expat pkgs/main/linux-64::expat-2.7.1-h6a678d5_0 2025-09-09T14:09:17.3294633Z ld_impl_linux-64 pkgs/main/linux-64::ld_impl_linux-64-2.40-h12ee557_0 2025-09-09T14:09:17.3295196Z libffi pkgs/main/linux-64::libffi-3.4.4-h6a678d5_1 2025-09-09T14:09:17.3295792Z libgcc-ng pkgs/main/linux-64::libgcc-ng-11.2.0-h1234567_1 2025-09-09T14:09:17.3296386Z libgomp pkgs/main/linux-64::libgomp-11.2.0-h1234567_1 2025-09-09T14:09:17.3296938Z libstdcxx-ng pkgs/main/linux-64::libstdcxx-ng-11.2.0-h1234567_1 2025-09-09T14:09:17.3297584Z libxcb pkgs/main/linux-64::libxcb-1.17.0-h9b100fa_0 2025-09-09T14:09:17.3298195Z ncurses pkgs/main/linux-64::ncurses-6.5-h7934f7d_0 2025-09-09T14:09:17.3298628Z openssl pkgs/main/linux-64::openssl-3.0.17-h5eee18b_0 2025-09-09T14:09:17.3299191Z pip pkgs/main/noarch::pip-25.2-pyhc872135_0 2025-09-09T14:09:17.3299714Z pthread-stubs pkgs/main/linux-64::pthread-stubs-0.3-h0ce48e5_1 2025-09-09T14:09:17.3300367Z python pkgs/main/linux-64::python-3.9.23-he99959a_0 2025-09-09T14:09:17.3300965Z readline pkgs/main/linux-64::readline-8.3-hc2a1206_0 2025-09-09T14:09:17.3301606Z setuptools pkgs/main/linux-64::setuptools-78.1.1-py39h06a4308_0 2025-09-09T14:09:17.3302124Z sqlite pkgs/main/linux-64::sqlite-3.50.2-hb25bd0a_1 2025-09-09T14:09:17.3302592Z tk pkgs/main/linux-64::tk-8.6.15-h54e0aa7_0 2025-09-09T14:09:17.3303078Z tzdata pkgs/main/noarch::tzdata-2025b-h04d1e81_0 2025-09-09T14:09:17.3303659Z wheel pkgs/main/linux-64::wheel-0.45.1-py39h06a4308_0 2025-09-09T14:09:17.3304198Z xorg-libx11 pkgs/main/linux-64::xorg-libx11-1.8.12-h9b100fa_1 2025-09-09T14:09:17.3304919Z xorg-libxau pkgs/main/linux-64::xorg-libxau-1.0.12-h9b100fa_0 2025-09-09T14:09:17.3305603Z xorg-libxdmcp pkgs/main/linux-64::xorg-libxdmcp-1.1.5-h9b100fa_0 2025-09-09T14:09:17.3306220Z xorg-xorgproto pkgs/main/linux-64::xorg-xorgproto-2024.1-h5eee18b_1 2025-09-09T14:09:17.3306792Z xz pkgs/main/linux-64::xz-5.6.4-h5eee18b_1 2025-09-09T14:09:17.3307198Z zlib pkgs/main/linux-64::zlib-1.2.13-h5eee18b_1 2025-09-09T14:09:17.3307667Z 2025-09-09T14:09:17.3307674Z 2025-09-09T14:09:17.3307679Z 2025-09-09T14:09:17.3307829Z Downloading and Extracting Packages 2025-09-09T14:09:17.3308104Z 2025-09-09T14:09:17.3308303Z libxcb-1.17.0 | 430 KB | : 0% 0/1 [00:00=4.10.0 (from torch) 2025-09-09T14:09:27.3148952Z Downloading https://download.pytorch.org/whl/nightly/typing_extensions-4.14.1-py3-none-any.whl.metadata (3.0 kB) 2025-09-09T14:09:27.3149534Z Collecting sympy>=1.13.3 (from torch) 2025-09-09T14:09:27.3150052Z Downloading https://download.pytorch.org/whl/nightly/sympy-1.14.0-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:09:27.3150585Z Collecting networkx>=2.5.1 (from torch) 2025-09-09T14:09:27.3151107Z Downloading https://download.pytorch.org/whl/nightly/networkx-3.5-py3-none-any.whl.metadata (6.3 kB) 2025-09-09T14:09:27.3151641Z Collecting jinja2 (from torch) 2025-09-09T14:09:27.3152134Z Downloading https://download.pytorch.org/whl/nightly/jinja2-3.1.6-py3-none-any.whl.metadata (2.9 kB) 2025-09-09T14:09:27.3152672Z Collecting fsspec>=0.8.5 (from torch) 2025-09-09T14:09:27.3153197Z Downloading https://download.pytorch.org/whl/nightly/fsspec-2025.7.0-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:09:27.3153775Z Collecting nvidia-cuda-nvrtc-cu12==12.6.77 (from torch) 2025-09-09T14:09:27.3154477Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cuda_nvrtc_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl (23.7 MB) 2025-09-09T14:09:27.3155766Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/23.7 MB ? eta -:--:-- 2025-09-09T14:09:27.3156620Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 23.7/23.7 MB 212.4 MB/s 0:00:00 2025-09-09T14:09:27.3157178Z [?25hCollecting nvidia-cuda-runtime-cu12==12.6.77 (from torch) 2025-09-09T14:09:27.3158021Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cuda_runtime_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl (897 kB) 2025-09-09T14:09:27.3158898Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/897.7 kB ? eta -:--:-- 2025-09-09T14:09:27.3159537Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 897.7/897.7 kB 89.0 MB/s 0:00:00 2025-09-09T14:09:27.3160077Z [?25hCollecting nvidia-cuda-cupti-cu12==12.6.80 (from torch) 2025-09-09T14:09:27.3160795Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cuda_cupti_cu12-12.6.80-py3-none-manylinux2014_x86_64.whl (8.9 MB) 2025-09-09T14:09:27.3161688Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/8.9 MB ? eta -:--:-- 2025-09-09T14:09:27.3162309Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.9/8.9 MB 209.1 MB/s 0:00:00 2025-09-09T14:09:33.9820559Z [?25hCollecting nvidia-cudnn-cu12==9.10.2.21 (from torch) 2025-09-09T14:09:33.9821465Z Downloading https://download.pytorch.org/whl/nightly/nvidia_cudnn_cu12-9.10.2.21-py3-none-manylinux_2_27_x86_64.whl.metadata (1.8 kB) 2025-09-09T14:09:33.9822154Z Collecting nvidia-cublas-cu12==12.6.4.1 (from torch) 2025-09-09T14:09:33.9822940Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cublas_cu12-12.6.4.1-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (393.1 MB) 2025-09-09T14:09:33.9823925Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/393.1 MB ? eta -:--:-- 2025-09-09T14:09:33.9824654Z  ━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 76.8/393.1 MB 385.0 MB/s eta 0:00:01 2025-09-09T14:09:33.9825410Z  ━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━ 153.9/393.1 MB 383.4 MB/s eta 0:00:01 2025-09-09T14:09:33.9826134Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━ 231.5/393.1 MB 383.8 MB/s eta 0:00:01 2025-09-09T14:09:33.9826846Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━ 308.8/393.1 MB 384.5 MB/s eta 0:00:01 2025-09-09T14:09:33.9827545Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺ 386.4/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9828261Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9828937Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9829866Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9830542Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9831460Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9832131Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9832803Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9833466Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9834144Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9834845Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9835537Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9836302Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 393.0/393.1 MB 384.2 MB/s eta 0:00:01 2025-09-09T14:09:33.9836948Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 393.1/393.1 MB 83.1 MB/s 0:00:03 2025-09-09T14:09:33.9837476Z [?25hCollecting nvidia-cufft-cu12==11.3.0.4 (from torch) 2025-09-09T14:09:33.9838170Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cufft_cu12-11.3.0.4-py3-none-manylinux2014_x86_64.whl (200.2 MB) 2025-09-09T14:09:33.9839025Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/200.2 MB ? eta -:--:-- 2025-09-09T14:09:33.9839718Z  ━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━ 84.4/200.2 MB 423.8 MB/s eta 0:00:01 2025-09-09T14:09:33.9840438Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━ 169.9/200.2 MB 423.6 MB/s eta 0:00:01 2025-09-09T14:09:33.9841123Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 422.8 MB/s eta 0:00:01 2025-09-09T14:09:33.9841791Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 422.8 MB/s eta 0:00:01 2025-09-09T14:09:33.9842444Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 200.0/200.2 MB 422.8 MB/s eta 0:00:01 2025-09-09T14:09:33.9843090Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 200.2/200.2 MB 168.1 MB/s 0:00:01 2025-09-09T14:09:33.9843712Z [?25hCollecting nvidia-curand-cu12==10.3.7.77 (from torch) 2025-09-09T14:09:33.9844416Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_curand_cu12-10.3.7.77-py3-none-manylinux2014_x86_64.whl (56.3 MB) 2025-09-09T14:09:33.9845354Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/56.3 MB ? eta -:--:-- 2025-09-09T14:09:33.9845991Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 56.1/56.3 MB 399.9 MB/s eta 0:00:01 2025-09-09T14:09:33.9846628Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 56.3/56.3 MB 213.3 MB/s 0:00:00 2025-09-09T14:09:33.9847152Z [?25hCollecting nvidia-cusolver-cu12==11.7.1.2 (from torch) 2025-09-09T14:09:33.9847864Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cusolver_cu12-11.7.1.2-py3-none-manylinux2014_x86_64.whl (158.2 MB) 2025-09-09T14:09:33.9848727Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/158.2 MB ? eta -:--:-- 2025-09-09T14:09:33.9849401Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━ 86.8/158.2 MB 435.1 MB/s eta 0:00:01 2025-09-09T14:09:33.9850109Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 158.1/158.2 MB 436.1 MB/s eta 0:00:01 2025-09-09T14:09:33.9850768Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 158.1/158.2 MB 436.1 MB/s eta 0:00:01 2025-09-09T14:09:33.9851427Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 158.1/158.2 MB 436.1 MB/s eta 0:00:01 2025-09-09T14:09:33.9852089Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 158.1/158.2 MB 436.1 MB/s eta 0:00:01 2025-09-09T14:09:38.4920044Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 158.2/158.2 MB 145.0 MB/s 0:00:01 2025-09-09T14:09:38.4920776Z [?25hCollecting nvidia-cusparse-cu12==12.5.4.2 (from torch) 2025-09-09T14:09:38.4921772Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cusparse_cu12-12.5.4.2-py3-none-manylinux2014_x86_64.whl (216.6 MB) 2025-09-09T14:09:38.4922677Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/216.6 MB ? eta -:--:-- 2025-09-09T14:09:38.4923365Z  ━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━ 84.4/216.6 MB 423.2 MB/s eta 0:00:01 2025-09-09T14:09:38.4924078Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━ 169.9/216.6 MB 423.6 MB/s eta 0:00:01 2025-09-09T14:09:38.4924774Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 423.2 MB/s eta 0:00:01 2025-09-09T14:09:38.4925451Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 423.2 MB/s eta 0:00:01 2025-09-09T14:09:38.4926388Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 423.2 MB/s eta 0:00:01 2025-09-09T14:09:38.4927213Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 423.2 MB/s eta 0:00:01 2025-09-09T14:09:38.4927890Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 216.5/216.6 MB 423.2 MB/s eta 0:00:01 2025-09-09T14:09:38.4928553Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 216.6/216.6 MB 142.6 MB/s 0:00:01 2025-09-09T14:09:38.4929096Z [?25hCollecting nvidia-cusparselt-cu12==0.7.1 (from torch) 2025-09-09T14:09:38.4929848Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cusparselt_cu12-0.7.1-py3-none-manylinux2014_x86_64.whl.metadata (7.0 kB) 2025-09-09T14:09:38.4930573Z Collecting nvidia-nccl-cu12==2.27.5 (from torch) 2025-09-09T14:09:38.4931360Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.0 kB) 2025-09-09T14:09:38.4932146Z Collecting nvidia-nvshmem-cu12==3.3.20 (from torch) 2025-09-09T14:09:38.4932934Z Downloading https://download.pytorch.org/whl/nightly/nvidia_nvshmem_cu12-3.3.20-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (2.1 kB) 2025-09-09T14:09:38.4933690Z Collecting nvidia-nvtx-cu12==12.6.77 (from torch) 2025-09-09T14:09:38.4934362Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_nvtx_cu12-12.6.77-py3-none-manylinux2014_x86_64.whl (89 kB) 2025-09-09T14:09:38.4935040Z Collecting nvidia-nvjitlink-cu12==12.6.85 (from torch) 2025-09-09T14:09:38.4935829Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_nvjitlink_cu12-12.6.85-py3-none-manylinux2010_x86_64.manylinux_2_12_x86_64.whl (19.7 MB) 2025-09-09T14:09:38.4936777Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/19.7 MB ? eta -:--:-- 2025-09-09T14:09:38.4937401Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 19.7/19.7 MB 149.8 MB/s 0:00:00 2025-09-09T14:09:38.4937929Z [?25hCollecting nvidia-cufile-cu12==1.11.1.6 (from torch) 2025-09-09T14:09:38.4938742Z Downloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (1.5 kB) 2025-09-09T14:09:38.4939556Z Collecting pytorch-triton==3.4.0+gitf7888497 (from torch) 2025-09-09T14:09:38.4940381Z Downloading https://download.pytorch.org/whl/nightly/pytorch_triton-3.4.0%2Bgitf7888497-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (1.8 kB) 2025-09-09T14:09:38.4941575Z Requirement already satisfied: setuptools>=40.8.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from pytorch-triton==3.4.0+gitf7888497->torch) (78.1.1) 2025-09-09T14:09:38.4942435Z Collecting importlib-metadata (from pytorch-triton==3.4.0+gitf7888497->torch) 2025-09-09T14:09:38.4943136Z Downloading https://download.pytorch.org/whl/nightly/importlib_metadata-7.1.0-py3-none-any.whl (24 kB) 2025-09-09T14:09:38.4944039Z INFO: pip is looking at multiple versions of networkx to determine which version is compatible with other requirements. This could take a while. 2025-09-09T14:09:38.4944691Z Collecting networkx>=2.5.1 (from torch) 2025-09-09T14:09:38.4945208Z Downloading https://download.pytorch.org/whl/nightly/networkx-3.2.1-py3-none-any.whl (1.6 MB) 2025-09-09T14:09:38.4945941Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.6 MB ? eta -:--:-- 2025-09-09T14:09:38.4946542Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 108.1 MB/s 0:00:00 2025-09-09T14:09:38.4947079Z [?25hCollecting mpmath<1.4,>=1.1.0 (from sympy>=1.13.3->torch) 2025-09-09T14:09:38.4947668Z Downloading https://download.pytorch.org/whl/nightly/mpmath-1.3.0-py3-none-any.whl (536 kB) 2025-09-09T14:09:38.4948537Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/536.2 kB ? eta -:--:-- 2025-09-09T14:09:38.4949164Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 536.2/536.2 kB 48.7 MB/s 0:00:00 2025-09-09T14:09:38.4949909Z [?25hCollecting zipp>=0.5 (from importlib-metadata->pytorch-triton==3.4.0+gitf7888497->torch) 2025-09-09T14:09:38.4950596Z Downloading https://download.pytorch.org/whl/nightly/zipp-3.19.2-py3-none-any.whl (9.0 kB) 2025-09-09T14:09:38.4951125Z Collecting MarkupSafe>=2.0 (from jinja2->torch) 2025-09-09T14:09:38.4951867Z Downloading https://download.pytorch.org/whl/nightly/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.0 kB) 2025-09-09T14:09:38.4952943Z Downloading https://download.pytorch.org/whl/nightly/cu126/torch-2.9.0.dev20250825%2Bcu126-cp39-cp39-manylinux_2_28_x86_64.whl (831.8 MB) 2025-09-09T14:09:38.4953831Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/831.8 MB ? eta -:--:-- 2025-09-09T14:09:38.4954513Z  ━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 82.6/831.8 MB 413.1 MB/s eta 0:00:02 2025-09-09T14:09:38.4955259Z  ━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 170.4/831.8 MB 424.8 MB/s eta 0:00:02 2025-09-09T14:09:38.4955977Z  ━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━ 244.8/831.8 MB 406.3 MB/s eta 0:00:02 2025-09-09T14:09:38.4956766Z  ━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━ 332.7/831.8 MB 414.8 MB/s eta 0:00:02 2025-09-09T14:09:38.4957476Z  ━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 420.2/831.8 MB 414.4 MB/s eta 0:00:01 2025-09-09T14:09:38.4958195Z  ━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━ 507.8/831.8 MB 435.7 MB/s eta 0:00:01 2025-09-09T14:09:38.4958984Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━ 595.3/831.8 MB 435.3 MB/s eta 0:00:01 2025-09-09T14:09:38.4959713Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━ 680.5/831.8 MB 430.5 MB/s eta 0:00:01 2025-09-09T14:09:45.9241664Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━ 767.6/831.8 MB 429.5 MB/s eta 0:00:01 2025-09-09T14:09:45.9242477Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9243196Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9243902Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9245184Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9246030Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9246700Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9247356Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9248021Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9248682Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9249354Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9250018Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9250688Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9251353Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9252022Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9252678Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9253339Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9254096Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9254842Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9255521Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9256175Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9256833Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9257491Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9258162Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9258825Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9259495Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9260157Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9260814Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9261466Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9262124Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9262842Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 831.8/831.8 MB 430.0 MB/s eta 0:00:01 2025-09-09T14:09:45.9263483Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 831.8/831.8 MB 39.2 MB/s 0:00:08 2025-09-09T14:09:45.9264585Z [?25hDownloading https://download.pytorch.org/whl/nightly/nvidia_cudnn_cu12-9.10.2.21-py3-none-manylinux_2_27_x86_64.whl (706.8 MB) 2025-09-09T14:09:45.9265405Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/706.8 MB ? eta -:--:-- 2025-09-09T14:09:45.9266070Z  ━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 72.9/706.8 MB 365.3 MB/s eta 0:00:02 2025-09-09T14:09:45.9266783Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 160.7/706.8 MB 400.8 MB/s eta 0:00:02 2025-09-09T14:09:45.9267789Z  ━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━ 248.3/706.8 MB 411.9 MB/s eta 0:00:02 2025-09-09T14:09:45.9268491Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━ 335.8/706.8 MB 435.7 MB/s eta 0:00:01 2025-09-09T14:09:45.9269323Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━ 423.9/706.8 MB 436.0 MB/s eta 0:00:01 2025-09-09T14:09:52.7879841Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━ 511.4/706.8 MB 436.1 MB/s eta 0:00:01 2025-09-09T14:09:52.7880845Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━ 599.0/706.8 MB 436.3 MB/s eta 0:00:01 2025-09-09T14:09:52.7881763Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 687.1/706.8 MB 436.6 MB/s eta 0:00:01 2025-09-09T14:09:52.7882504Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7883177Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7883875Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7884544Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7885220Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7885901Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7886556Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7887243Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7887935Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7888598Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7889309Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7889976Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7890644Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7891654Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7892306Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7893131Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7893796Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7894448Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7895105Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7895767Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7896451Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7897128Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7897789Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7898455Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7899112Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 706.7/706.8 MB 436.2 MB/s eta 0:00:01 2025-09-09T14:09:52.7899739Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 706.8/706.8 MB 46.2 MB/s 0:00:06 2025-09-09T14:09:52.7900816Z [?25hDownloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cufile_cu12-1.11.1.6-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (1.1 MB) 2025-09-09T14:09:52.7901741Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.1 MB ? eta -:--:-- 2025-09-09T14:09:52.7902431Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.1/1.1 MB 6.9 MB/s 0:00:00 2025-09-09T14:09:52.7903296Z [?25hDownloading https://download.pytorch.org/whl/nightly/cu126/nvidia_cusparselt_cu12-0.7.1-py3-none-manylinux2014_x86_64.whl (287.2 MB) 2025-09-09T14:09:52.7904149Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/287.2 MB ? eta -:--:-- 2025-09-09T14:09:52.7904817Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 85.2/287.2 MB 426.6 MB/s eta 0:00:01 2025-09-09T14:09:52.7905519Z  ━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━ 170.7/287.2 MB 425.7 MB/s eta 0:00:01 2025-09-09T14:09:52.7906238Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 256.6/287.2 MB 425.7 MB/s eta 0:00:01 2025-09-09T14:09:52.7906936Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:52.7907642Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4571828Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4572547Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4573386Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4574191Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4575158Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4576046Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4576711Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4577477Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4578136Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4579141Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 287.0/287.2 MB 426.5 MB/s eta 0:00:01 2025-09-09T14:09:59.4579777Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 287.2/287.2 MB 84.7 MB/s 0:00:03 2025-09-09T14:09:59.4580911Z [?25hDownloading https://download.pytorch.org/whl/nightly/cu126/nvidia_nccl_cu12-2.27.5-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (322.3 MB) 2025-09-09T14:09:59.4581845Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/322.3 MB ? eta -:--:-- 2025-09-09T14:09:59.4582514Z  ━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 84.1/322.3 MB 421.3 MB/s eta 0:00:01 2025-09-09T14:09:59.4583223Z  ━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━ 168.6/322.3 MB 419.2 MB/s eta 0:00:01 2025-09-09T14:09:59.4583925Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━ 253.8/322.3 MB 420.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4584628Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4585301Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4585960Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4586619Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4587284Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4587940Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4588610Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4589270Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4589945Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4590603Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4591261Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 322.2/322.3 MB 421.0 MB/s eta 0:00:01 2025-09-09T14:09:59.4591894Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 322.3/322.3 MB 91.7 MB/s 0:00:03 2025-09-09T14:09:59.4592815Z [?25hDownloading https://download.pytorch.org/whl/nightly/nvidia_nvshmem_cu12-3.3.20-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (124.7 MB) 2025-09-09T14:09:59.4593829Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/124.7 MB ? eta -:--:-- 2025-09-09T14:09:59.4594574Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━ 84.1/124.7 MB 421.6 MB/s eta 0:00:01 2025-09-09T14:09:59.4595256Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 124.5/124.7 MB 419.4 MB/s eta 0:00:01 2025-09-09T14:09:59.4595967Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 124.5/124.7 MB 419.4 MB/s eta 0:00:01 2025-09-09T14:09:59.4596685Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 124.5/124.7 MB 419.4 MB/s eta 0:00:01 2025-09-09T14:09:59.4597326Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 124.7/124.7 MB 138.6 MB/s 0:00:00 2025-09-09T14:09:59.4598324Z [?25hDownloading https://download.pytorch.org/whl/nightly/pytorch_triton-3.4.0%2Bgitf7888497-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (155.5 MB) 2025-09-09T14:09:59.4599273Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/155.5 MB ? eta -:--:-- 2025-09-09T14:09:59.4599931Z  ━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━ 84.9/155.5 MB 425.2 MB/s eta 0:00:01 2025-09-09T14:09:59.4600607Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 155.5/155.5 MB 424.1 MB/s eta 0:00:01 2025-09-09T14:09:59.4601273Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 155.5/155.5 MB 424.1 MB/s eta 0:00:01 2025-09-09T14:10:07.8315422Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 155.5/155.5 MB 424.1 MB/s eta 0:00:01 2025-09-09T14:10:07.8316429Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 155.5/155.5 MB 424.1 MB/s eta 0:00:01 2025-09-09T14:10:07.8317210Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 155.5/155.5 MB 424.1 MB/s eta 0:00:01 2025-09-09T14:10:07.8317954Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 155.5/155.5 MB 125.0 MB/s 0:00:01 2025-09-09T14:10:07.8318692Z [?25hDownloading https://download.pytorch.org/whl/nightly/fsspec-2025.7.0-py3-none-any.whl (199 kB) 2025-09-09T14:10:07.8319413Z Downloading https://download.pytorch.org/whl/nightly/sympy-1.14.0-py3-none-any.whl (6.3 MB) 2025-09-09T14:10:07.8320119Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/6.3 MB ? eta -:--:-- 2025-09-09T14:10:07.8320714Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.3/6.3 MB 203.3 MB/s 0:00:00 2025-09-09T14:10:07.8321775Z [?25hDownloading https://download.pytorch.org/whl/nightly/typing_extensions-4.14.1-py3-none-any.whl (43 kB) 2025-09-09T14:10:07.8322551Z Downloading https://download.pytorch.org/whl/nightly/filelock-3.19.1-py3-none-any.whl (15 kB) 2025-09-09T14:10:07.8323250Z Downloading https://download.pytorch.org/whl/nightly/jinja2-3.1.6-py3-none-any.whl (134 kB) 2025-09-09T14:10:07.8324282Z Downloading https://download.pytorch.org/whl/nightly/MarkupSafe-3.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (20 kB) 2025-09-09T14:10:07.8326585Z Installing collected packages: nvidia-cusparselt-cu12, mpmath, zipp, typing-extensions, sympy, nvidia-nvtx-cu12, nvidia-nvshmem-cu12, nvidia-nvjitlink-cu12, nvidia-nccl-cu12, nvidia-curand-cu12, nvidia-cufile-cu12, nvidia-cuda-runtime-cu12, nvidia-cuda-nvrtc-cu12, nvidia-cuda-cupti-cu12, nvidia-cublas-cu12, networkx, MarkupSafe, fsspec, filelock, nvidia-cusparse-cu12, nvidia-cufft-cu12, nvidia-cudnn-cu12, jinja2, importlib-metadata, pytorch-triton, nvidia-cusolver-cu12, torch 2025-09-09T14:10:07.8328540Z [?25l 2025-09-09T14:10:07.8328933Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8329476Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8330029Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8330565Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8331094Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8331630Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8332162Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8332699Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8333245Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8333780Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8334331Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8334856Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8335392Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  0/27 [nvidia-cusparselt-cu12] 2025-09-09T14:10:07.8335931Z  ━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  1/27 [mpmath] 2025-09-09T14:10:07.8336471Z  ━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  1/27 [mpmath] 2025-09-09T14:10:07.8337008Z  ━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  2/27 [zipp] 2025-09-09T14:10:07.8337656Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8338195Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8338821Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8339364Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8339908Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8340436Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8340974Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8341504Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8342063Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8342598Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8343146Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8343684Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8344212Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8344755Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8345294Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8345824Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:07.8346470Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0514946Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0516063Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0516873Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0517427Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0517973Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0518513Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0519044Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0519607Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0520136Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0520692Z  ━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/27 [sympy] 2025-09-09T14:10:15.0521280Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0521888Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0522496Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0523093Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0523697Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0524314Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0524914Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  6/27 [nvidia-nvshmem-cu12] 2025-09-09T14:10:15.0525545Z  ━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/27 [nvidia-nvjitlink-cu12] 2025-09-09T14:10:15.0526148Z  ━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/27 [nvidia-nvjitlink-cu12] 2025-09-09T14:10:15.0526747Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0527337Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0527911Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0528734Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0529315Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0529979Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0530562Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0531141Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0531767Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0532344Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0532921Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0533519Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0534109Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0534692Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0535269Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0535845Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0536424Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0536998Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/27 [nvidia-nccl-cu12] 2025-09-09T14:10:15.0537692Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  9/27 [nvidia-curand-cu12] 2025-09-09T14:10:15.0538288Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  9/27 [nvidia-curand-cu12] 2025-09-09T14:10:15.0538995Z  ━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━ 11/27 [nvidia-cuda-runtime-cu12] 2025-09-09T14:10:15.0539618Z  ━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━ 12/27 [nvidia-cuda-nvrtc-cu12] 2025-09-09T14:10:15.0540225Z  ━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━ 12/27 [nvidia-cuda-nvrtc-cu12] 2025-09-09T14:10:15.0540836Z  ━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━ 13/27 [nvidia-cuda-cupti-cu12] 2025-09-09T14:10:22.2611726Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2612402Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2613045Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2613669Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2614264Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2614875Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2615469Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2616074Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2616696Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2617285Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2617901Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2618492Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2619096Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2619696Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2620293Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2620897Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2621739Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2622339Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2623115Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2623712Z  ━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━ 14/27 [nvidia-cublas-cu12] 2025-09-09T14:10:22.2624283Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━ 15/27 [networkx] 2025-09-09T14:10:22.2624828Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━ 15/27 [networkx] 2025-09-09T14:10:22.2625382Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━ 15/27 [networkx] 2025-09-09T14:10:22.2626001Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━ 15/27 [networkx] 2025-09-09T14:10:22.2626542Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━ 15/27 [networkx] 2025-09-09T14:10:22.2627110Z  ━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━ 17/27 [fsspec] 2025-09-09T14:10:22.2627688Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2628299Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2628915Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2629514Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2630118Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2630836Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2631442Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2632143Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2632744Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2633349Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2633946Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━ 19/27 [nvidia-cusparse-cu12] 2025-09-09T14:10:22.2634547Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:22.2635155Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:22.2635738Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:22.2636423Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:22.2637005Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:22.2644514Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:29.4644507Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:29.4645153Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:29.4645817Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:29.4646416Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━ 20/27 [nvidia-cufft-cu12] 2025-09-09T14:10:29.4647041Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4647642Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4648234Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4648844Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4649432Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4650034Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4651004Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4651604Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4652370Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4652965Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4653579Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4654179Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4654782Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4655391Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4655981Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4656581Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4657175Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4657767Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4658349Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4658937Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4659524Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4660227Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4660962Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4661550Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4662136Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4662721Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4663314Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4664098Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4664707Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4665294Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4665898Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4666491Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4667083Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4667671Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4668255Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4668851Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4669437Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4670035Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━ 21/27 [nvidia-cudnn-cu12] 2025-09-09T14:10:29.4670634Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━ 22/27 [jinja2] 2025-09-09T14:10:36.8638764Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━ 23/27 [importlib-metadata] 2025-09-09T14:10:36.8639465Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8640052Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8640621Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8641533Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8642101Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8642854Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8643440Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8644007Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8644579Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8645171Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8645811Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8646395Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8646986Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━ 24/27 [pytorch-triton] 2025-09-09T14:10:36.8647587Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8648192Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8648794Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8649397Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8649992Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8650708Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8651407Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8652016Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8652618Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 25/27 [nvidia-cusolver-cu12] 2025-09-09T14:10:36.8653172Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8653699Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8654216Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8654762Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8655287Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8655822Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8656340Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8656852Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8657374Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8657899Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8658410Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8658950Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8659463Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8659996Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8660507Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8661024Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8661540Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8662055Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8662570Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8663219Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:36.8663974Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6011910Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6012515Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6013030Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6013549Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6014068Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6014580Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6015121Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6015632Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6016167Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6016691Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6017208Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6017729Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6018240Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6018763Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6019471Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6019984Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6020512Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6021122Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6021647Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6022172Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6022690Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6023216Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6023754Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6024279Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6024794Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6025335Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6025854Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6026365Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6026879Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6027386Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6027923Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6028445Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6028957Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6029497Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6030033Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6030577Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6031101Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6031614Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6032139Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6032792Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6033308Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6033920Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6034438Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6034957Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6035467Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:44.6035984Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6494487Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6495153Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6495697Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6496287Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6496810Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6497327Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6497834Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6498348Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6498859Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6499652Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6500160Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6500916Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6501441Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6501952Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6502472Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6502983Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6503500Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6504044Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6504554Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6505095Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6505604Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6506166Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6506687Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6507201Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6507715Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6508250Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6508779Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 26/27 [torch] 2025-09-09T14:10:53.6509290Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 27/27 [torch] 2025-09-09T14:10:53.6509629Z [?25h 2025-09-09T14:10:53.6512489Z Successfully installed MarkupSafe-3.0.2 filelock-3.19.1 fsspec-2025.7.0 importlib-metadata-7.1.0 jinja2-3.1.6 mpmath-1.3.0 networkx-3.2.1 nvidia-cublas-cu12-12.6.4.1 nvidia-cuda-cupti-cu12-12.6.80 nvidia-cuda-nvrtc-cu12-12.6.77 nvidia-cuda-runtime-cu12-12.6.77 nvidia-cudnn-cu12-9.10.2.21 nvidia-cufft-cu12-11.3.0.4 nvidia-cufile-cu12-1.11.1.6 nvidia-curand-cu12-10.3.7.77 nvidia-cusolver-cu12-11.7.1.2 nvidia-cusparse-cu12-12.5.4.2 nvidia-cusparselt-cu12-0.7.1 nvidia-nccl-cu12-2.27.5 nvidia-nvjitlink-cu12-12.6.85 nvidia-nvshmem-cu12-3.3.20 nvidia-nvtx-cu12-12.6.77 pytorch-triton-3.4.0+gitf7888497 sympy-1.14.0 torch-2.9.0.dev20250825+cu126 typing-extensions-4.14.1 zipp-3.19.2 2025-09-09T14:10:53.6516750Z WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. 2025-09-09T14:10:53.6518230Z + pip install -r dev-requirements.txt 2025-09-09T14:10:53.6518612Z Collecting pytest (from -r dev-requirements.txt (line 2)) 2025-09-09T14:10:53.6519053Z Downloading pytest-8.4.2-py3-none-any.whl.metadata (7.7 kB) 2025-09-09T14:10:53.6519560Z Collecting unittest-xml-reporting (from -r dev-requirements.txt (line 3)) 2025-09-09T14:10:53.6520135Z Downloading unittest_xml_reporting-3.2.0-py2.py3-none-any.whl.metadata (11 kB) 2025-09-09T14:10:53.6520696Z Collecting parameterized (from -r dev-requirements.txt (line 4)) 2025-09-09T14:10:53.6521206Z Downloading parameterized-0.9.0-py2.py3-none-any.whl.metadata (18 kB) 2025-09-09T14:10:53.6521711Z Collecting packaging (from -r dev-requirements.txt (line 5)) 2025-09-09T14:10:53.6522199Z Downloading packaging-25.0-py3-none-any.whl.metadata (3.3 kB) 2025-09-09T14:10:53.6522678Z Collecting transformers (from -r dev-requirements.txt (line 6)) 2025-09-09T14:10:53.6523163Z Downloading transformers-4.56.1-py3-none-any.whl.metadata (42 kB) 2025-09-09T14:10:53.6523651Z Collecting hypothesis (from -r dev-requirements.txt (line 7)) 2025-09-09T14:10:53.6524128Z Downloading hypothesis-6.138.15-py3-none-any.whl.metadata (5.6 kB) 2025-09-09T14:10:53.6524626Z Collecting sentencepiece (from -r dev-requirements.txt (line 8)) 2025-09-09T14:10:53.6525246Z Downloading sentencepiece-0.2.1-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (10 kB) 2025-09-09T14:10:53.6525872Z Collecting expecttest (from -r dev-requirements.txt (line 9)) 2025-09-09T14:10:53.6526353Z Downloading expecttest-0.3.0-py3-none-any.whl.metadata (3.8 kB) 2025-09-09T14:10:53.6526836Z Collecting bitsandbytes (from -r dev-requirements.txt (line 12)) 2025-09-09T14:10:53.6527379Z Downloading bitsandbytes-0.47.0-py3-none-manylinux_2_24_x86_64.whl.metadata (11 kB) 2025-09-09T14:10:53.6527918Z Collecting matplotlib (from -r dev-requirements.txt (line 13)) 2025-09-09T14:10:53.6528512Z Downloading matplotlib-3.9.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (11 kB) 2025-09-09T14:10:53.6529098Z Collecting pandas (from -r dev-requirements.txt (line 14)) 2025-09-09T14:10:53.6529653Z Downloading pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (91 kB) 2025-09-09T14:10:53.6530212Z Collecting fire (from -r dev-requirements.txt (line 15)) 2025-09-09T14:10:53.6530630Z Downloading fire-0.7.1-py3-none-any.whl.metadata (5.8 kB) 2025-09-09T14:10:53.6531076Z Collecting tabulate (from -r dev-requirements.txt (line 16)) 2025-09-09T14:10:53.6531533Z Downloading tabulate-0.9.0-py3-none-any.whl.metadata (34 kB) 2025-09-09T14:10:53.6531986Z Collecting tiktoken (from -r dev-requirements.txt (line 17)) 2025-09-09T14:10:53.6532562Z Downloading tiktoken-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.7 kB) 2025-09-09T14:10:53.6533139Z Collecting blobfile (from -r dev-requirements.txt (line 18)) 2025-09-09T14:11:00.1311774Z Downloading blobfile-3.1.0-py3-none-any.whl.metadata (15 kB) 2025-09-09T14:11:00.1312380Z Collecting lm_eval (from -r dev-requirements.txt (line 19)) 2025-09-09T14:11:00.1312978Z Downloading lm_eval-0.4.9.1-py3-none-any.whl.metadata (53 kB) 2025-09-09T14:11:00.1313594Z Collecting diskcache (from -r dev-requirements.txt (line 21)) 2025-09-09T14:11:00.1314089Z Downloading diskcache-5.6.3-py3-none-any.whl.metadata (20 kB) 2025-09-09T14:11:00.1314559Z Collecting pycocotools (from -r dev-requirements.txt (line 22)) 2025-09-09T14:11:00.1315681Z Downloading pycocotools-2.0.10-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (1.3 kB) 2025-09-09T14:11:00.1316268Z Collecting tqdm (from -r dev-requirements.txt (line 23)) 2025-09-09T14:11:00.1316745Z Downloading tqdm-4.67.1-py3-none-any.whl.metadata (57 kB) 2025-09-09T14:11:00.1317687Z Requirement already satisfied: importlib_metadata in /opt/conda/envs/venv/lib/python3.9/site-packages (from -r dev-requirements.txt (line 24)) (7.1.0) 2025-09-09T14:11:00.1318457Z Collecting ninja (from -r dev-requirements.txt (line 27)) 2025-09-09T14:11:00.1319018Z Downloading ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (5.1 kB) 2025-09-09T14:11:00.1319607Z Collecting cmake<4.0.0,>=3.19.0 (from -r dev-requirements.txt (line 30)) 2025-09-09T14:11:00.1320189Z Downloading cmake-3.31.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.3 kB) 2025-09-09T14:11:00.1320750Z Collecting ruff==0.11.6 (from -r dev-requirements.txt (line 33)) 2025-09-09T14:11:00.1321316Z Downloading ruff-0.11.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (25 kB) 2025-09-09T14:11:00.1321904Z Collecting pre-commit (from -r dev-requirements.txt (line 34)) 2025-09-09T14:11:00.1322382Z Downloading pre_commit-4.3.0-py2.py3-none-any.whl.metadata (1.2 kB) 2025-09-09T14:11:00.1322931Z Collecting exceptiongroup>=1 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:00.1323474Z Downloading exceptiongroup-1.3.0-py3-none-any.whl.metadata (6.7 kB) 2025-09-09T14:11:00.1323998Z Collecting iniconfig>=1 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:00.1324493Z Downloading iniconfig-2.1.0-py3-none-any.whl.metadata (2.7 kB) 2025-09-09T14:11:00.1324986Z Collecting pluggy<2,>=1.5 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:00.1325465Z Downloading pluggy-1.6.0-py3-none-any.whl.metadata (4.8 kB) 2025-09-09T14:11:00.1325944Z Collecting pygments>=2.7.2 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:00.1326452Z Downloading pygments-2.19.2-py3-none-any.whl.metadata (2.5 kB) 2025-09-09T14:11:00.1326925Z Collecting tomli>=1 (from pytest->-r dev-requirements.txt (line 2)) 2025-09-09T14:11:00.1327381Z Downloading tomli-2.2.1-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:00.1327898Z Collecting lxml (from unittest-xml-reporting->-r dev-requirements.txt (line 3)) 2025-09-09T14:11:00.1328527Z Downloading lxml-6.0.1-cp39-cp39-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl.metadata (3.8 kB) 2025-09-09T14:11:00.1329443Z Requirement already satisfied: filelock in /opt/conda/envs/venv/lib/python3.9/site-packages (from transformers->-r dev-requirements.txt (line 6)) (3.19.1) 2025-09-09T14:11:00.1330372Z Collecting huggingface-hub<1.0,>=0.34.0 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1330951Z Downloading huggingface_hub-0.34.4-py3-none-any.whl.metadata (14 kB) 2025-09-09T14:11:00.1331483Z Collecting numpy>=1.17 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1332082Z Downloading numpy-2.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB) 2025-09-09T14:11:00.1332694Z Collecting pyyaml>=5.1 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1333296Z Downloading PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (2.1 kB) 2025-09-09T14:11:00.1333927Z Collecting regex!=2019.12.17 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1334620Z Downloading regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (40 kB) 2025-09-09T14:11:00.1335305Z Collecting requests (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1335810Z Downloading requests-2.32.5-py3-none-any.whl.metadata (4.9 kB) 2025-09-09T14:11:00.1336339Z Collecting tokenizers<=0.23.0,>=0.22.0 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1337004Z Downloading tokenizers-0.22.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (6.8 kB) 2025-09-09T14:11:00.1337797Z Collecting safetensors>=0.4.3 (from transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1338463Z Downloading safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.1 kB) 2025-09-09T14:11:00.1339588Z Requirement already satisfied: fsspec>=2023.5.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) (2025.7.0) 2025-09-09T14:11:00.1340958Z Requirement already satisfied: typing-extensions>=3.7.4.3 in /opt/conda/envs/venv/lib/python3.9/site-packages (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) (4.14.1) 2025-09-09T14:11:00.1342022Z Collecting hf-xet<2.0.0,>=1.1.3 (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:00.1342716Z Downloading hf_xet-1.1.9-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (4.7 kB) 2025-09-09T14:11:00.1343313Z Collecting attrs>=22.2.0 (from hypothesis->-r dev-requirements.txt (line 7)) 2025-09-09T14:11:00.1343794Z Downloading attrs-25.3.0-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:00.1344325Z Collecting sortedcontainers<3.0.0,>=2.1.0 (from hypothesis->-r dev-requirements.txt (line 7)) 2025-09-09T14:11:00.1344938Z Downloading sortedcontainers-2.4.0-py2.py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:00.1345857Z Requirement already satisfied: torch<3,>=2.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from bitsandbytes->-r dev-requirements.txt (line 12)) (2.9.0.dev20250825+cu126) 2025-09-09T14:11:00.1347084Z Requirement already satisfied: sympy>=1.13.3 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (1.14.0) 2025-09-09T14:11:00.1348286Z Requirement already satisfied: networkx>=2.5.1 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.2.1) 2025-09-09T14:11:00.1349481Z Requirement already satisfied: jinja2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.1.6) 2025-09-09T14:11:00.1350801Z Requirement already satisfied: nvidia-cuda-nvrtc-cu12==12.6.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.77) 2025-09-09T14:11:00.1352154Z Requirement already satisfied: nvidia-cuda-runtime-cu12==12.6.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.77) 2025-09-09T14:11:00.1353496Z Requirement already satisfied: nvidia-cuda-cupti-cu12==12.6.80 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.80) 2025-09-09T14:11:00.1354822Z Requirement already satisfied: nvidia-cudnn-cu12==9.10.2.21 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (9.10.2.21) 2025-09-09T14:11:00.1356148Z Requirement already satisfied: nvidia-cublas-cu12==12.6.4.1 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.4.1) 2025-09-09T14:11:00.1357507Z Requirement already satisfied: nvidia-cufft-cu12==11.3.0.4 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (11.3.0.4) 2025-09-09T14:11:00.1358827Z Requirement already satisfied: nvidia-curand-cu12==10.3.7.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (10.3.7.77) 2025-09-09T14:11:00.1360171Z Requirement already satisfied: nvidia-cusolver-cu12==11.7.1.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (11.7.1.2) 2025-09-09T14:11:00.1361531Z Requirement already satisfied: nvidia-cusparse-cu12==12.5.4.2 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.5.4.2) 2025-09-09T14:11:00.1363028Z Requirement already satisfied: nvidia-cusparselt-cu12==0.7.1 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (0.7.1) 2025-09-09T14:11:00.1364678Z Requirement already satisfied: nvidia-nccl-cu12==2.27.5 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (2.27.5) 2025-09-09T14:11:00.1365969Z Requirement already satisfied: nvidia-nvshmem-cu12==3.3.20 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.3.20) 2025-09-09T14:11:00.1367275Z Requirement already satisfied: nvidia-nvtx-cu12==12.6.77 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.77) 2025-09-09T14:11:00.1368601Z Requirement already satisfied: nvidia-nvjitlink-cu12==12.6.85 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (12.6.85) 2025-09-09T14:11:00.1369939Z Requirement already satisfied: nvidia-cufile-cu12==1.11.1.6 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (1.11.1.6) 2025-09-09T14:11:11.4790421Z Requirement already satisfied: pytorch-triton==3.4.0+gitf7888497 in /opt/conda/envs/venv/lib/python3.9/site-packages (from torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.4.0+gitf7888497) 2025-09-09T14:11:11.4794073Z Requirement already satisfied: setuptools>=40.8.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from pytorch-triton==3.4.0+gitf7888497->torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (78.1.1) 2025-09-09T14:11:11.4796949Z Collecting contourpy>=1.0.1 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4798007Z Downloading contourpy-1.3.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (5.4 kB) 2025-09-09T14:11:11.4798886Z Collecting cycler>=0.10 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4799397Z Downloading cycler-0.12.1-py3-none-any.whl.metadata (3.8 kB) 2025-09-09T14:11:11.4799908Z Collecting fonttools>=4.22.0 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4800553Z Downloading fonttools-4.59.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (109 kB) 2025-09-09T14:11:11.4801204Z Collecting kiwisolver>=1.3.1 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4801850Z Downloading kiwisolver-1.4.7-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl.metadata (6.3 kB) 2025-09-09T14:11:11.4802481Z Collecting pillow>=8 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4803098Z Downloading pillow-11.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (9.0 kB) 2025-09-09T14:11:11.4803728Z Collecting pyparsing>=2.3.1 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4804256Z Downloading pyparsing-3.2.3-py3-none-any.whl.metadata (5.0 kB) 2025-09-09T14:11:11.4804793Z Collecting python-dateutil>=2.7 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4805404Z Downloading python_dateutil-2.9.0.post0-py2.py3-none-any.whl.metadata (8.4 kB) 2025-09-09T14:11:11.4806028Z Collecting importlib-resources>=3.2.0 (from matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:11.4806625Z Downloading importlib_resources-6.5.2-py3-none-any.whl.metadata (3.9 kB) 2025-09-09T14:11:11.4807188Z Collecting pytz>=2020.1 (from pandas->-r dev-requirements.txt (line 14)) 2025-09-09T14:11:11.4807700Z Downloading pytz-2025.2-py2.py3-none-any.whl.metadata (22 kB) 2025-09-09T14:11:11.4808196Z Collecting tzdata>=2022.7 (from pandas->-r dev-requirements.txt (line 14)) 2025-09-09T14:11:11.4808947Z Downloading tzdata-2025.2-py2.py3-none-any.whl.metadata (1.4 kB) 2025-09-09T14:11:11.4809442Z Collecting termcolor (from fire->-r dev-requirements.txt (line 15)) 2025-09-09T14:11:11.4809929Z Downloading termcolor-3.1.0-py3-none-any.whl.metadata (6.4 kB) 2025-09-09T14:11:11.4810599Z Collecting pycryptodomex>=3.8 (from blobfile->-r dev-requirements.txt (line 18)) 2025-09-09T14:11:11.4811278Z Downloading pycryptodomex-3.23.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (3.4 kB) 2025-09-09T14:11:11.4811927Z Collecting urllib3<3,>=1.25.3 (from blobfile->-r dev-requirements.txt (line 18)) 2025-09-09T14:11:11.4812430Z Downloading urllib3-2.5.0-py3-none-any.whl.metadata (6.5 kB) 2025-09-09T14:11:11.4812925Z Collecting accelerate>=0.26.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4813444Z Downloading accelerate-1.10.1-py3-none-any.whl.metadata (19 kB) 2025-09-09T14:11:11.4813940Z Collecting evaluate (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4814430Z Downloading evaluate-0.4.5-py3-none-any.whl.metadata (9.5 kB) 2025-09-09T14:11:11.4814933Z Collecting datasets<4.0,>=2.16.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4815428Z Downloading datasets-3.6.0-py3-none-any.whl.metadata (19 kB) 2025-09-09T14:11:11.4815921Z Collecting jsonlines (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4816405Z Downloading jsonlines-4.0.0-py3-none-any.whl.metadata (1.6 kB) 2025-09-09T14:11:11.4816888Z Collecting numexpr (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4817482Z Downloading numexpr-2.10.2-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl.metadata (8.1 kB) 2025-09-09T14:11:11.4818073Z Collecting peft>=0.2.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4818536Z Downloading peft-0.17.1-py3-none-any.whl.metadata (14 kB) 2025-09-09T14:11:11.4819008Z Collecting pybind11>=2.6.2 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4819510Z Downloading pybind11-3.0.1-py3-none-any.whl.metadata (10.0 kB) 2025-09-09T14:11:11.4820007Z Collecting pytablewriter (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4820534Z Downloading pytablewriter-1.2.1-py3-none-any.whl.metadata (38 kB) 2025-09-09T14:11:11.4821067Z Collecting rouge-score>=0.0.4 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4821500Z Downloading rouge_score-0.1.2.tar.gz (17 kB) 2025-09-09T14:11:11.4822110Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:11:11.4822644Z [?25hCollecting sacrebleu>=1.5.0 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4823167Z Downloading sacrebleu-2.5.1-py3-none-any.whl.metadata (51 kB) 2025-09-09T14:11:11.4823676Z Collecting scikit-learn>=0.24.1 (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4824330Z Downloading scikit_learn-1.6.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (18 kB) 2025-09-09T14:11:11.4824965Z Collecting sqlitedict (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4825387Z Downloading sqlitedict-2.1.0.tar.gz (21 kB) 2025-09-09T14:11:11.4825789Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:11:11.4826327Z [?25hCollecting tqdm-multiprocess (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4826913Z Downloading tqdm_multiprocess-0.0.11-py3-none-any.whl.metadata (5.7 kB) 2025-09-09T14:11:11.4827494Z Collecting zstandard (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4828108Z Downloading zstandard-0.24.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl.metadata (3.1 kB) 2025-09-09T14:11:11.4828722Z Collecting dill (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4829167Z Downloading dill-0.4.0-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:11.4829643Z Collecting word2number (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4830155Z Downloading word2number-1.1.zip (9.7 kB) 2025-09-09T14:11:11.4830545Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:11:11.4831070Z [?25hCollecting more_itertools (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4831603Z Downloading more_itertools-10.8.0-py3-none-any.whl.metadata (39 kB) 2025-09-09T14:11:11.4832259Z Collecting pyarrow>=15.0.0 (from datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4832855Z Downloading pyarrow-21.0.0-cp39-cp39-manylinux_2_28_x86_64.whl.metadata (3.3 kB) 2025-09-09T14:11:11.4833374Z Collecting dill (from lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4833816Z Downloading dill-0.3.8-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:11.4834330Z Collecting xxhash (from datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4834986Z Downloading xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB) 2025-09-09T14:11:11.4835671Z Collecting multiprocess<0.70.17 (from datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4836285Z Downloading multiprocess-0.70.16-py39-none-any.whl.metadata (7.2 kB) 2025-09-09T14:11:11.4836980Z Collecting fsspec>=2023.5.0 (from huggingface-hub<1.0,>=0.34.0->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:11.4837629Z Downloading fsspec-2025.3.0-py3-none-any.whl.metadata (11 kB) 2025-09-09T14:11:11.4838305Z Collecting aiohttp!=4.0.0a0,!=4.0.0a1 (from fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4839082Z Downloading aiohttp-3.12.15-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (7.7 kB) 2025-09-09T14:11:11.4840021Z Requirement already satisfied: zipp>=0.5 in /opt/conda/envs/venv/lib/python3.9/site-packages (from importlib_metadata->-r dev-requirements.txt (line 24)) (3.19.2) 2025-09-09T14:11:11.4840851Z Collecting cfgv>=2.0.0 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:11.4841352Z Downloading cfgv-3.4.0-py2.py3-none-any.whl.metadata (8.5 kB) 2025-09-09T14:11:11.4841856Z Collecting identify>=1.0.0 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:11.4842375Z Downloading identify-2.6.14-py2.py3-none-any.whl.metadata (4.4 kB) 2025-09-09T14:11:11.4842895Z Collecting nodeenv>=0.11.1 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:11.4843408Z Downloading nodeenv-1.9.1-py2.py3-none-any.whl.metadata (21 kB) 2025-09-09T14:11:11.4843931Z Collecting virtualenv>=20.10.0 (from pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:11.4844458Z Downloading virtualenv-20.34.0-py3-none-any.whl.metadata (4.6 kB) 2025-09-09T14:11:11.4852369Z Collecting psutil (from accelerate>=0.26.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4853223Z Downloading psutil-7.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (22 kB) 2025-09-09T14:11:11.4854239Z Collecting aiohappyeyeballs>=2.5.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4855050Z Downloading aiohappyeyeballs-2.6.1-py3-none-any.whl.metadata (5.9 kB) 2025-09-09T14:11:11.4855820Z Collecting aiosignal>=1.4.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4856537Z Downloading aiosignal-1.4.0-py3-none-any.whl.metadata (3.7 kB) 2025-09-09T14:11:11.4857275Z Collecting async-timeout<6.0,>=4.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:11.4858028Z Downloading async_timeout-5.0.1-py3-none-any.whl.metadata (5.1 kB) 2025-09-09T14:11:11.4858760Z Collecting frozenlist>=1.1.1 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4292592Z Downloading frozenlist-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (18 kB) 2025-09-09T14:11:20.4294191Z Collecting multidict<7.0,>=4.5 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4295130Z Downloading multidict-6.6.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (5.3 kB) 2025-09-09T14:11:20.4296053Z Collecting propcache>=0.2.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4296904Z Downloading propcache-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (12 kB) 2025-09-09T14:11:20.4297728Z Collecting yarl<2.0,>=1.17.0 (from aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4298529Z Downloading yarl-1.20.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (73 kB) 2025-09-09T14:11:20.4299357Z Collecting idna>=2.0 (from yarl<2.0,>=1.17.0->aiohttp!=4.0.0a0,!=4.0.0a1->fsspec[http]<=2025.3.0,>=2023.1.0->datasets<4.0,>=2.16.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4300047Z Downloading idna-3.10-py3-none-any.whl.metadata (10 kB) 2025-09-09T14:11:20.4300595Z Collecting six>=1.5 (from python-dateutil>=2.7->matplotlib->-r dev-requirements.txt (line 13)) 2025-09-09T14:11:20.4301167Z Downloading six-1.17.0-py2.py3-none-any.whl.metadata (1.7 kB) 2025-09-09T14:11:20.4301745Z Collecting charset_normalizer<4,>=2 (from requests->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:20.4302579Z Downloading charset_normalizer-3.4.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl.metadata (36 kB) 2025-09-09T14:11:20.4303422Z Collecting certifi>=2017.4.17 (from requests->transformers->-r dev-requirements.txt (line 6)) 2025-09-09T14:11:20.4304023Z Downloading certifi-2025.8.3-py3-none-any.whl.metadata (2.4 kB) 2025-09-09T14:11:20.4304588Z Collecting absl-py (from rouge-score>=0.0.4->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4305120Z Downloading absl_py-2.3.1-py3-none-any.whl.metadata (3.3 kB) 2025-09-09T14:11:20.4305635Z Collecting nltk (from rouge-score>=0.0.4->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4306140Z Downloading nltk-3.9.1-py3-none-any.whl.metadata (2.9 kB) 2025-09-09T14:11:20.4306688Z Collecting portalocker (from sacrebleu>=1.5.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4307271Z Downloading portalocker-3.2.0-py3-none-any.whl.metadata (8.7 kB) 2025-09-09T14:11:20.4307813Z Collecting colorama (from sacrebleu>=1.5.0->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4308377Z Downloading colorama-0.4.6-py2.py3-none-any.whl.metadata (17 kB) 2025-09-09T14:11:20.4308939Z Collecting scipy>=1.6.0 (from scikit-learn>=0.24.1->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4309600Z Downloading scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl.metadata (60 kB) 2025-09-09T14:11:20.4310262Z Collecting joblib>=1.2.0 (from scikit-learn>=0.24.1->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4310801Z Downloading joblib-1.5.2-py3-none-any.whl.metadata (5.6 kB) 2025-09-09T14:11:20.4311378Z Collecting threadpoolctl>=3.1.0 (from scikit-learn>=0.24.1->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4311977Z Downloading threadpoolctl-3.6.0-py3-none-any.whl.metadata (13 kB) 2025-09-09T14:11:20.4312894Z Requirement already satisfied: mpmath<1.4,>=1.1.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from sympy>=1.13.3->torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (1.3.0) 2025-09-09T14:11:20.4313868Z Collecting distlib<1,>=0.3.7 (from virtualenv>=20.10.0->pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:20.4314608Z Downloading distlib-0.4.0-py2.py3-none-any.whl.metadata (5.2 kB) 2025-09-09T14:11:20.4315214Z Collecting platformdirs<5,>=3.9.1 (from virtualenv>=20.10.0->pre-commit->-r dev-requirements.txt (line 34)) 2025-09-09T14:11:20.4315893Z Downloading platformdirs-4.4.0-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:11:20.4316859Z Requirement already satisfied: MarkupSafe>=2.0 in /opt/conda/envs/venv/lib/python3.9/site-packages (from jinja2->torch<3,>=2.2->bitsandbytes->-r dev-requirements.txt (line 12)) (3.0.2) 2025-09-09T14:11:20.4317818Z Collecting click (from nltk->rouge-score>=0.0.4->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4318350Z Downloading click-8.1.8-py3-none-any.whl.metadata (2.3 kB) 2025-09-09T14:11:20.4318905Z Collecting DataProperty<2,>=1.1.0 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4319490Z Downloading DataProperty-1.1.0-py3-none-any.whl.metadata (11 kB) 2025-09-09T14:11:20.4320076Z Collecting mbstrdecoder<2,>=1.0.0 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4320658Z Downloading mbstrdecoder-1.1.4-py3-none-any.whl.metadata (4.3 kB) 2025-09-09T14:11:20.4321249Z Collecting pathvalidate<4,>=2.3.0 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4321837Z Downloading pathvalidate-3.3.1-py3-none-any.whl.metadata (12 kB) 2025-09-09T14:11:20.4322400Z Collecting tabledata<2,>=1.3.1 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4322961Z Downloading tabledata-1.3.4-py3-none-any.whl.metadata (3.7 kB) 2025-09-09T14:11:20.4323512Z Collecting tcolorpy<1,>=0.0.5 (from pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4324068Z Downloading tcolorpy-0.1.7-py3-none-any.whl.metadata (6.3 kB) 2025-09-09T14:11:20.4324686Z Collecting typepy<2,>=1.3.2 (from typepy[datetime]<2,>=1.3.2->pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4325301Z Downloading typepy-1.3.4-py3-none-any.whl.metadata (9.2 kB) 2025-09-09T14:11:20.4325910Z Collecting chardet<6,>=3.0.4 (from mbstrdecoder<2,>=1.0.0->pytablewriter->lm_eval->-r dev-requirements.txt (line 19)) 2025-09-09T14:11:20.4326522Z Downloading chardet-5.2.0-py3-none-any.whl.metadata (3.4 kB) 2025-09-09T14:11:20.4327065Z Downloading ruff-0.11.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (11.5 MB) 2025-09-09T14:11:20.4327882Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/11.5 MB ? eta -:--:-- 2025-09-09T14:11:20.4328558Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺ 11.3/11.5 MB 273.6 MB/s eta 0:00:01 2025-09-09T14:11:20.4329197Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.5/11.5 MB 39.6 MB/s 0:00:00 2025-09-09T14:11:20.4329865Z [?25hDownloading cmake-3.31.6-py3-none-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (27.8 MB) 2025-09-09T14:11:20.4330651Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/27.8 MB ? eta -:--:-- 2025-09-09T14:11:20.4331283Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 27.8/27.8 MB 419.6 MB/s eta 0:00:01 2025-09-09T14:11:20.4332036Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 27.8/27.8 MB 419.6 MB/s eta 0:00:01 2025-09-09T14:11:20.4332671Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 27.8/27.8 MB 55.4 MB/s 0:00:00 2025-09-09T14:11:20.4333224Z [?25hDownloading pytest-8.4.2-py3-none-any.whl (365 kB) 2025-09-09T14:11:20.4333611Z Downloading pluggy-1.6.0-py3-none-any.whl (20 kB) 2025-09-09T14:11:20.4334044Z Downloading unittest_xml_reporting-3.2.0-py2.py3-none-any.whl (20 kB) 2025-09-09T14:11:20.4334540Z Downloading parameterized-0.9.0-py2.py3-none-any.whl (20 kB) 2025-09-09T14:11:20.4334977Z Downloading packaging-25.0-py3-none-any.whl (66 kB) 2025-09-09T14:11:20.4335381Z Downloading transformers-4.56.1-py3-none-any.whl (11.6 MB) 2025-09-09T14:11:20.4335956Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/11.6 MB ? eta -:--:-- 2025-09-09T14:11:20.4336601Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 11.5/11.6 MB 325.2 MB/s eta 0:00:01 2025-09-09T14:11:20.4337227Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 11.6/11.6 MB 35.8 MB/s 0:00:00 2025-09-09T14:11:20.4337775Z [?25hDownloading huggingface_hub-0.34.4-py3-none-any.whl (561 kB) 2025-09-09T14:11:20.4338356Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/561.5 kB ? eta -:--:-- 2025-09-09T14:11:20.4338962Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 561.5/561.5 kB 3.2 MB/s 0:00:00 2025-09-09T14:11:20.4339622Z [?25hDownloading hf_xet-1.1.9-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.2 MB) 2025-09-09T14:11:25.1019270Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/3.2 MB ? eta -:--:-- 2025-09-09T14:11:25.1020155Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.2/3.2 MB 20.6 MB/s 0:00:00 2025-09-09T14:11:25.1021001Z [?25hDownloading tokenizers-0.22.0-cp39-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (3.3 MB) 2025-09-09T14:11:25.1021725Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/3.3 MB ? eta -:--:-- 2025-09-09T14:11:25.1022332Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 3.3/3.3 MB 22.6 MB/s 0:00:00 2025-09-09T14:11:25.1022875Z [?25hDownloading hypothesis-6.138.15-py3-none-any.whl (533 kB) 2025-09-09T14:11:25.1023475Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/533.6 kB ? eta -:--:-- 2025-09-09T14:11:25.1024431Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 533.6/533.6 kB 2.9 MB/s 0:00:00 2025-09-09T14:11:25.1025026Z [?25hDownloading sortedcontainers-2.4.0-py2.py3-none-any.whl (29 kB) 2025-09-09T14:11:25.1025666Z Downloading sentencepiece-0.2.1-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (1.4 MB) 2025-09-09T14:11:25.1026529Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.4 MB ? eta -:--:-- 2025-09-09T14:11:25.1027142Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.4/1.4 MB 11.3 MB/s 0:00:00 2025-09-09T14:11:25.1027663Z [?25hDownloading expecttest-0.3.0-py3-none-any.whl (8.2 kB) 2025-09-09T14:11:25.1028177Z Downloading bitsandbytes-0.47.0-py3-none-manylinux_2_24_x86_64.whl (61.3 MB) 2025-09-09T14:11:25.1028836Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/61.3 MB ? eta -:--:-- 2025-09-09T14:11:25.1029484Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 61.1/61.3 MB 426.0 MB/s eta 0:00:01 2025-09-09T14:11:25.1030183Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 61.1/61.3 MB 426.0 MB/s eta 0:00:01 2025-09-09T14:11:25.1030863Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 61.1/61.3 MB 426.0 MB/s eta 0:00:01 2025-09-09T14:11:25.1031502Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 61.3/61.3 MB 83.1 MB/s 0:00:00 2025-09-09T14:11:25.1032204Z [?25hDownloading matplotlib-3.9.4-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (8.3 MB) 2025-09-09T14:11:25.1032910Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/8.3 MB ? eta -:--:-- 2025-09-09T14:11:25.1033554Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺ 8.1/8.3 MB 437.0 MB/s eta 0:00:01 2025-09-09T14:11:25.1034185Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 8.3/8.3 MB 29.4 MB/s 0:00:00 2025-09-09T14:11:25.1034889Z [?25hDownloading pandas-2.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (12.4 MB) 2025-09-09T14:11:25.1035611Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/12.4 MB ? eta -:--:-- 2025-09-09T14:11:25.1036248Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 12.3/12.4 MB 425.8 MB/s eta 0:00:01 2025-09-09T14:11:25.1037133Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 12.4/12.4 MB 44.6 MB/s 0:00:00 2025-09-09T14:11:25.1037635Z [?25hDownloading fire-0.7.1-py3-none-any.whl (115 kB) 2025-09-09T14:11:25.1038012Z Downloading tabulate-0.9.0-py3-none-any.whl (35 kB) 2025-09-09T14:11:25.1038534Z Downloading tiktoken-0.11.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.2 MB) 2025-09-09T14:11:25.1039207Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.2 MB ? eta -:--:-- 2025-09-09T14:11:25.1039909Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 9.2 MB/s 0:00:00 2025-09-09T14:11:25.1040408Z [?25hDownloading blobfile-3.1.0-py3-none-any.whl (75 kB) 2025-09-09T14:11:25.1040886Z Downloading urllib3-2.5.0-py3-none-any.whl (129 kB) 2025-09-09T14:11:25.1041272Z Downloading lm_eval-0.4.9.1-py3-none-any.whl (7.5 MB) 2025-09-09T14:11:25.1041813Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/7.5 MB ? eta -:--:-- 2025-09-09T14:11:25.1042484Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 7.3/7.5 MB 275.7 MB/s eta 0:00:01 2025-09-09T14:11:25.1043120Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 7.5/7.5 MB 33.9 MB/s 0:00:00 2025-09-09T14:11:25.1043638Z [?25hDownloading datasets-3.6.0-py3-none-any.whl (491 kB) 2025-09-09T14:11:25.1044042Z Downloading dill-0.3.8-py3-none-any.whl (116 kB) 2025-09-09T14:11:25.1044415Z Downloading fsspec-2025.3.0-py3-none-any.whl (193 kB) 2025-09-09T14:11:25.1044838Z Downloading multiprocess-0.70.16-py39-none-any.whl (133 kB) 2025-09-09T14:11:25.1045254Z Downloading diskcache-5.6.3-py3-none-any.whl (45 kB) 2025-09-09T14:11:25.1045810Z Downloading pycocotools-2.0.10-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (453 kB) 2025-09-09T14:11:25.1046335Z Downloading tqdm-4.67.1-py3-none-any.whl (78 kB) 2025-09-09T14:11:25.1046828Z Downloading ninja-1.13.0-py3-none-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (180 kB) 2025-09-09T14:11:25.1047357Z Downloading pre_commit-4.3.0-py2.py3-none-any.whl (220 kB) 2025-09-09T14:11:25.1047771Z Downloading accelerate-1.10.1-py3-none-any.whl (374 kB) 2025-09-09T14:11:25.1048284Z Downloading numpy-2.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (19.5 MB) 2025-09-09T14:11:25.1048950Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/19.5 MB ? eta -:--:-- 2025-09-09T14:11:25.1049599Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 19.4/19.5 MB 424.9 MB/s eta 0:00:01 2025-09-09T14:11:25.1050227Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 19.5/19.5 MB 50.4 MB/s 0:00:00 2025-09-09T14:11:25.1050924Z [?25hDownloading aiohttp-3.12.15-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (1.6 MB) 2025-09-09T14:11:25.1051612Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.6 MB ? eta -:--:-- 2025-09-09T14:11:25.1052196Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 11.9 MB/s 0:00:00 2025-09-09T14:11:25.1052723Z [?25hDownloading async_timeout-5.0.1-py3-none-any.whl (6.2 kB) 2025-09-09T14:11:25.1053344Z Downloading multidict-6.6.4-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (239 kB) 2025-09-09T14:11:25.1054162Z Downloading yarl-1.20.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (327 kB) 2025-09-09T14:11:25.1054705Z Downloading aiohappyeyeballs-2.6.1-py3-none-any.whl (15 kB) 2025-09-09T14:11:25.1055130Z Downloading aiosignal-1.4.0-py3-none-any.whl (7.5 kB) 2025-09-09T14:11:25.1055509Z Downloading attrs-25.3.0-py3-none-any.whl (63 kB) 2025-09-09T14:11:25.1055958Z Downloading cfgv-3.4.0-py2.py3-none-any.whl (7.2 kB) 2025-09-09T14:11:25.1056477Z Downloading contourpy-1.3.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (321 kB) 2025-09-09T14:11:28.3150996Z Downloading cycler-0.12.1-py3-none-any.whl (8.3 kB) 2025-09-09T14:11:28.3151491Z Downloading evaluate-0.4.5-py3-none-any.whl (84 kB) 2025-09-09T14:11:28.3152054Z Downloading exceptiongroup-1.3.0-py3-none-any.whl (16 kB) 2025-09-09T14:11:28.3152780Z Downloading fonttools-4.59.2-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (4.8 MB) 2025-09-09T14:11:28.3153563Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/4.8 MB ? eta -:--:-- 2025-09-09T14:11:28.3154244Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 4.8/4.8 MB 32.5 MB/s 0:00:00 2025-09-09T14:11:28.3155155Z [?25hDownloading frozenlist-1.7.0-cp39-cp39-manylinux_2_5_x86_64.manylinux1_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (225 kB) 2025-09-09T14:11:28.3155884Z Downloading identify-2.6.14-py2.py3-none-any.whl (99 kB) 2025-09-09T14:11:28.3156284Z Downloading idna-3.10-py3-none-any.whl (70 kB) 2025-09-09T14:11:28.3156806Z Downloading importlib_resources-6.5.2-py3-none-any.whl (37 kB) 2025-09-09T14:11:28.3157247Z Downloading iniconfig-2.1.0-py3-none-any.whl (6.0 kB) 2025-09-09T14:11:28.3157828Z Downloading kiwisolver-1.4.7-cp39-cp39-manylinux_2_12_x86_64.manylinux2010_x86_64.whl (1.6 MB) 2025-09-09T14:11:28.3158550Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.6 MB ? eta -:--:-- 2025-09-09T14:11:28.3159147Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.6/1.6 MB 11.2 MB/s 0:00:00 2025-09-09T14:11:28.3159828Z [?25hDownloading lxml-6.0.1-cp39-cp39-manylinux_2_26_x86_64.manylinux_2_28_x86_64.whl (5.3 MB) 2025-09-09T14:11:28.3160554Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/5.3 MB ? eta -:--:-- 2025-09-09T14:11:28.3161183Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.3/5.3 MB 32.6 MB/s 0:00:00 2025-09-09T14:11:28.3161727Z [?25hDownloading nodeenv-1.9.1-py2.py3-none-any.whl (22 kB) 2025-09-09T14:11:28.3162136Z Downloading peft-0.17.1-py3-none-any.whl (504 kB) 2025-09-09T14:11:28.3162659Z Downloading pillow-11.3.0-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (6.6 MB) 2025-09-09T14:11:28.3163338Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/6.6 MB ? eta -:--:-- 2025-09-09T14:11:28.3164080Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.6/6.6 MB 35.4 MB/s 0:00:00 2025-09-09T14:11:28.3164801Z [?25hDownloading propcache-0.3.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (200 kB) 2025-09-09T14:11:28.3165418Z Downloading pyarrow-21.0.0-cp39-cp39-manylinux_2_28_x86_64.whl (42.7 MB) 2025-09-09T14:11:28.3166057Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/42.7 MB ? eta -:--:-- 2025-09-09T14:11:28.3166714Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 42.7/42.7 MB 420.8 MB/s eta 0:00:01 2025-09-09T14:11:28.3167408Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 42.7/42.7 MB 420.8 MB/s eta 0:00:01 2025-09-09T14:11:28.3168140Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 42.7/42.7 MB 420.8 MB/s eta 0:00:01 2025-09-09T14:11:28.3168771Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 42.7/42.7 MB 61.8 MB/s 0:00:00 2025-09-09T14:11:28.3169569Z [?25hDownloading pybind11-3.0.1-py3-none-any.whl (293 kB) 2025-09-09T14:11:28.3170137Z Downloading pycryptodomex-3.23.0-cp37-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (2.3 MB) 2025-09-09T14:11:28.3170867Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/2.3 MB ? eta -:--:-- 2025-09-09T14:11:28.3171681Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 2.3/2.3 MB 16.0 MB/s 0:00:00 2025-09-09T14:11:28.3172199Z [?25hDownloading pygments-2.19.2-py3-none-any.whl (1.2 MB) 2025-09-09T14:11:28.3172764Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.2 MB ? eta -:--:-- 2025-09-09T14:11:28.3173347Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.2/1.2 MB 8.7 MB/s 0:00:00 2025-09-09T14:11:28.3173852Z [?25hDownloading pyparsing-3.2.3-py3-none-any.whl (111 kB) 2025-09-09T14:11:28.3174339Z Downloading python_dateutil-2.9.0.post0-py2.py3-none-any.whl (229 kB) 2025-09-09T14:11:28.3174813Z Downloading pytz-2025.2-py2.py3-none-any.whl (509 kB) 2025-09-09T14:11:28.3175323Z Downloading PyYAML-6.0.2-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (737 kB) 2025-09-09T14:11:28.3176015Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/737.4 kB ? eta -:--:-- 2025-09-09T14:11:28.3176622Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 737.4/737.4 kB 50.3 MB/s 0:00:00 2025-09-09T14:11:28.3177396Z [?25hDownloading regex-2025.9.1-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (789 kB) 2025-09-09T14:11:28.3178154Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/789.5 kB ? eta -:--:-- 2025-09-09T14:11:28.3178768Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 789.5/789.5 kB 53.9 MB/s 0:00:00 2025-09-09T14:11:28.3179294Z [?25hDownloading requests-2.32.5-py3-none-any.whl (64 kB) 2025-09-09T14:11:28.3179958Z Downloading charset_normalizer-3.4.3-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.manylinux_2_28_x86_64.whl (152 kB) 2025-09-09T14:11:28.3180602Z Downloading certifi-2025.8.3-py3-none-any.whl (161 kB) 2025-09-09T14:11:28.3180996Z Downloading sacrebleu-2.5.1-py3-none-any.whl (104 kB) 2025-09-09T14:11:28.3181550Z Downloading safetensors-0.6.2-cp38-abi3-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (485 kB) 2025-09-09T14:11:28.3182227Z Downloading scikit_learn-1.6.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (13.5 MB) 2025-09-09T14:11:28.3182920Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/13.5 MB ? eta -:--:-- 2025-09-09T14:11:28.3183527Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 13.5/13.5 MB 186.3 MB/s 0:00:00 2025-09-09T14:11:28.3184036Z [?25hDownloading joblib-1.5.2-py3-none-any.whl (308 kB) 2025-09-09T14:11:28.3184550Z Downloading scipy-1.13.1-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (38.6 MB) 2025-09-09T14:11:28.3185323Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/38.6 MB ? eta -:--:-- 2025-09-09T14:11:28.3186041Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 38.5/38.6 MB 411.9 MB/s eta 0:00:01 2025-09-09T14:11:28.3186676Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 38.6/38.6 MB 158.9 MB/s 0:00:00 2025-09-09T14:11:28.3187183Z [?25hDownloading six-1.17.0-py2.py3-none-any.whl (11 kB) 2025-09-09T14:11:28.3187600Z Downloading threadpoolctl-3.6.0-py3-none-any.whl (18 kB) 2025-09-09T14:11:28.3187987Z Downloading tomli-2.2.1-py3-none-any.whl (14 kB) 2025-09-09T14:11:28.3188370Z Downloading tzdata-2025.2-py2.py3-none-any.whl (347 kB) 2025-09-09T14:11:28.3188774Z Downloading virtualenv-20.34.0-py3-none-any.whl (6.0 MB) 2025-09-09T14:11:28.3189325Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/6.0 MB ? eta -:--:-- 2025-09-09T14:11:28.3189933Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 6.0/6.0 MB 170.5 MB/s 0:00:00 2025-09-09T14:11:32.6718684Z [?25hDownloading distlib-0.4.0-py2.py3-none-any.whl (469 kB) 2025-09-09T14:11:32.6719191Z Downloading platformdirs-4.4.0-py3-none-any.whl (18 kB) 2025-09-09T14:11:32.6719729Z Downloading absl_py-2.3.1-py3-none-any.whl (135 kB) 2025-09-09T14:11:32.6720246Z Downloading colorama-0.4.6-py2.py3-none-any.whl (25 kB) 2025-09-09T14:11:32.6720762Z Downloading jsonlines-4.0.0-py3-none-any.whl (8.7 kB) 2025-09-09T14:11:32.6721221Z Downloading more_itertools-10.8.0-py3-none-any.whl (69 kB) 2025-09-09T14:11:32.6721609Z Downloading nltk-3.9.1-py3-none-any.whl (1.5 MB) 2025-09-09T14:11:32.6722191Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/1.5 MB ? eta -:--:-- 2025-09-09T14:11:32.6722800Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 1.5/1.5 MB 129.4 MB/s 0:00:00 2025-09-09T14:11:32.6723312Z [?25hDownloading click-8.1.8-py3-none-any.whl (98 kB) 2025-09-09T14:11:32.6723832Z Downloading numexpr-2.10.2-cp39-cp39-manylinux_2_27_x86_64.manylinux_2_28_x86_64.whl (396 kB) 2025-09-09T14:11:32.6724367Z Downloading portalocker-3.2.0-py3-none-any.whl (22 kB) 2025-09-09T14:11:32.6725032Z Downloading psutil-7.0.0-cp36-abi3-manylinux_2_12_x86_64.manylinux2010_x86_64.manylinux_2_17_x86_64.manylinux2014_x86_64.whl (277 kB) 2025-09-09T14:11:32.6725705Z Downloading pytablewriter-1.2.1-py3-none-any.whl (91 kB) 2025-09-09T14:11:32.6726117Z Downloading DataProperty-1.1.0-py3-none-any.whl (27 kB) 2025-09-09T14:11:32.6726534Z Downloading mbstrdecoder-1.1.4-py3-none-any.whl (7.9 kB) 2025-09-09T14:11:32.6726919Z Downloading chardet-5.2.0-py3-none-any.whl (199 kB) 2025-09-09T14:11:32.6727321Z Downloading pathvalidate-3.3.1-py3-none-any.whl (24 kB) 2025-09-09T14:11:32.6727716Z Downloading tabledata-1.3.4-py3-none-any.whl (11 kB) 2025-09-09T14:11:32.6728380Z Downloading tcolorpy-0.1.7-py3-none-any.whl (8.1 kB) 2025-09-09T14:11:32.6728756Z Downloading typepy-1.3.4-py3-none-any.whl (31 kB) 2025-09-09T14:11:32.6729137Z Downloading termcolor-3.1.0-py3-none-any.whl (7.7 kB) 2025-09-09T14:11:32.6729570Z Downloading tqdm_multiprocess-0.0.11-py3-none-any.whl (9.8 kB) 2025-09-09T14:11:32.6730257Z Downloading xxhash-3.5.0-cp39-cp39-manylinux_2_17_x86_64.manylinux2014_x86_64.whl (193 kB) 2025-09-09T14:11:32.6730968Z Downloading zstandard-0.24.0-cp39-cp39-manylinux2014_x86_64.manylinux_2_17_x86_64.whl (5.6 MB) 2025-09-09T14:11:32.6731675Z [?25l ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 0.0/5.6 MB ? eta -:--:-- 2025-09-09T14:11:32.6732268Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 5.6/5.6 MB 195.2 MB/s 0:00:00 2025-09-09T14:11:32.6732868Z [?25hBuilding wheels for collected packages: rouge-score, sqlitedict, word2number 2025-09-09T14:11:32.6735044Z  DEPRECATION: Building 'rouge-score' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'rouge-score'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:11:32.6736904Z  Building wheel for rouge-score (setup.py) ... [?25l- done 2025-09-09T14:11:32.6737819Z [?25h Created wheel for rouge-score: filename=rouge_score-0.1.2-py3-none-any.whl size=24988 sha256=b226d5e762b9bb077ab1e2f8c33f482666d6da63e09f71cbf4495f437a86e51e 2025-09-09T14:11:32.6738760Z Stored in directory: /root/.cache/pip/wheels/9b/3d/39/09558097d3119ca0a4d462df68f22c6f3c1b345ac63a09b86e 2025-09-09T14:11:32.6740994Z  DEPRECATION: Building 'sqlitedict' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'sqlitedict'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:11:32.6742828Z  Building wheel for sqlitedict (setup.py) ... [?25l- done 2025-09-09T14:11:32.6743741Z [?25h Created wheel for sqlitedict: filename=sqlitedict-2.1.0-py3-none-any.whl size=16958 sha256=e66a50571390e610e8755ad0649d9abd589d333f37b90fbeaa4b46ba59950ad8 2025-09-09T14:11:32.6744689Z Stored in directory: /root/.cache/pip/wheels/f6/48/c4/942f7a1d556fddd2348cb9ac262f251873dfd8a39afec5678e 2025-09-09T14:11:32.6746921Z  DEPRECATION: Building 'word2number' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'word2number'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:11:32.6748747Z  Building wheel for word2number (setup.py) ... [?25l- done 2025-09-09T14:11:32.6749660Z [?25h Created wheel for word2number: filename=word2number-1.1-py3-none-any.whl size=5658 sha256=4061164156c4a2f9dec741260d4b738dcc7f1a276e084ee84b5a3777a796d133 2025-09-09T14:11:32.6750612Z Stored in directory: /root/.cache/pip/wheels/a0/4a/5b/d2f2df5c344ddbecb8bea759872c207ea91d93f57fb54e816e 2025-09-09T14:11:32.6751204Z Successfully built rouge-score sqlitedict word2number 2025-09-09T14:11:32.6755663Z Installing collected packages: word2number, sqlitedict, sortedcontainers, pytz, distlib, zstandard, xxhash, urllib3, tzdata, tqdm, tomli, threadpoolctl, termcolor, tcolorpy, tabulate, six, sentencepiece, safetensors, ruff, regex, pyyaml, pyparsing, pygments, pycryptodomex, pybind11, pyarrow, psutil, propcache, portalocker, pluggy, platformdirs, pillow, pathvalidate, parameterized, packaging, numpy, nodeenv, ninja, multidict, more_itertools, lxml, kiwisolver, joblib, iniconfig, importlib-resources, idna, identify, hf-xet, fsspec, frozenlist, fonttools, expecttest, exceptiongroup, diskcache, dill, cycler, colorama, cmake, click, charset_normalizer, chardet, cfgv, certifi, attrs, async-timeout, aiohappyeyeballs, absl-py, yarl, virtualenv, unittest-xml-reporting, tqdm-multiprocess, scipy, sacrebleu, requests, python-dateutil, pytest, pycocotools, numexpr, nltk, multiprocess, mbstrdecoder, jsonlines, hypothesis, fire, contourpy, blobfile, aiosignal, typepy, tiktoken, scikit-learn, rouge-score, pre-commit, pandas, matplotlib, huggingface-hub, aiohttp, tokenizers, bitsandbytes, accelerate, transformers, datasets, DataProperty, tabledata, peft, evaluate, pytablewriter, lm_eval 2025-09-09T14:11:32.6760377Z [?25l 2025-09-09T14:11:32.6760804Z  ━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  4/107 [distlib] 2025-09-09T14:11:32.6761359Z  ━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  7/107 [urllib3] 2025-09-09T14:11:32.6761915Z  ━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  8/107 [tzdata] 2025-09-09T14:11:32.6762463Z  ━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  18/107 [ruff] 2025-09-09T14:11:32.6763006Z  ━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  19/107 [regex] 2025-09-09T14:11:32.6763557Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  22/107 [pygments] 2025-09-09T14:11:32.6764550Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  22/107 [pygments] 2025-09-09T14:11:32.6765102Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  22/107 [pygments] 2025-09-09T14:11:32.6765684Z  ━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  22/107 [pygments] 2025-09-09T14:11:32.6766253Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  23/107 [pycryptodomex] 2025-09-09T14:11:32.6766849Z  ━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  23/107 [pycryptodomex] 2025-09-09T14:11:39.4310926Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  25/107 [pyarrow] 2025-09-09T14:11:39.4311552Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  25/107 [pyarrow] 2025-09-09T14:11:39.4312129Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  25/107 [pyarrow] 2025-09-09T14:11:39.4312695Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  25/107 [pyarrow] 2025-09-09T14:11:39.4313249Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  25/107 [pyarrow] 2025-09-09T14:11:39.4314901Z  ━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  25/107 [pyarrow] 2025-09-09T14:11:39.4315460Z  ━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━  26/107 [psutil] 2025-09-09T14:11:39.4316200Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  31/107 [pillow] 2025-09-09T14:11:39.4316832Z  ━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━━━━  31/107 [pillow] 2025-09-09T14:11:39.4317386Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4317943Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4318488Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4319040Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4319611Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4320155Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4320720Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4321258Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4321802Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4322352Z  ━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━━━━━━━━━  35/107 [numpy] 2025-09-09T14:11:39.4322894Z  ━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━━  40/107 [lxml] 2025-09-09T14:11:39.4323442Z  ━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━━  42/107 [joblib] 2025-09-09T14:11:39.4323999Z  ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/107 [idna] 2025-09-09T14:11:39.4324419Z  Attempting uninstall: fsspec 2025-09-09T14:11:39.4324975Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/107 [idna] 2025-09-09T14:11:39.4325437Z  Found existing installation: fsspec 2025.7.0 2025-09-09T14:11:39.4325932Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/107 [idna] 2025-09-09T14:11:39.4326348Z  Uninstalling fsspec-2025.7.0: 2025-09-09T14:11:39.4326809Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/107 [idna] 2025-09-09T14:11:39.4327248Z  Successfully uninstalled fsspec-2025.7.0 2025-09-09T14:11:39.4327744Z ━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━━  45/107 [idna] 2025-09-09T14:11:39.4328284Z  ━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━━  48/107 [fsspec] 2025-09-09T14:11:39.4328949Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━  50/107 [fonttools] 2025-09-09T14:11:39.4329512Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━  50/107 [fonttools] 2025-09-09T14:11:39.4330087Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━  50/107 [fonttools] 2025-09-09T14:11:39.4330645Z  ━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━━━━━━━  50/107 [fonttools] 2025-09-09T14:11:39.4331191Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4331733Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4332284Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4332847Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4333389Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4333927Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4334495Z  ━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━━  57/107 [cmake] 2025-09-09T14:11:39.4335047Z  ━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━━━━  60/107 [chardet] 2025-09-09T14:11:39.4335597Z  ━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━━━  66/107 [absl-py] 2025-09-09T14:11:39.4336157Z  ━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━━━━  68/107 [virtualenv] 2025-09-09T14:11:39.4336700Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:39.4337353Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:39.4337894Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0632307Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0632955Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0633504Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0634042Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0634582Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0635113Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0635687Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0636221Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0636856Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0637391Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0637920Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0638463Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0639001Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━━━  71/107 [scipy] 2025-09-09T14:11:47.0639545Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━━  75/107 [pytest] 2025-09-09T14:11:47.0640115Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━━  77/107 [numexpr] 2025-09-09T14:11:47.0640648Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━  78/107 [nltk] 2025-09-09T14:11:47.0641308Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━  78/107 [nltk] 2025-09-09T14:11:47.0641845Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━━━  78/107 [nltk] 2025-09-09T14:11:47.0642398Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━━  79/107 [multiprocess] 2025-09-09T14:11:47.0642971Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━━━  82/107 [hypothesis] 2025-09-09T14:11:47.0643506Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━━━  83/107 [fire] 2025-09-09T14:11:47.0644044Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━━━  88/107 [tiktoken] 2025-09-09T14:11:47.0644804Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0645373Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0645962Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0646528Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0647097Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0647666Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0648227Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0648793Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0649372Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━━  89/107 [scikit-learn] 2025-09-09T14:11:47.0649932Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0650494Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0651030Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0651568Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0652101Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0652641Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0653290Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0653823Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0654453Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0654987Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0655526Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0656065Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0656611Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:47.0657152Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5212736Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5213496Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5214197Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5214729Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5215268Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5215809Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━━  92/107 [pandas] 2025-09-09T14:11:54.5216357Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  93/107 [matplotlib] 2025-09-09T14:11:54.5216919Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  93/107 [matplotlib] 2025-09-09T14:11:54.5217491Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  93/107 [matplotlib] 2025-09-09T14:11:54.5218048Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  93/107 [matplotlib] 2025-09-09T14:11:54.5218869Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  93/107 [matplotlib] 2025-09-09T14:11:54.5219429Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━━  93/107 [matplotlib] 2025-09-09T14:11:54.5220004Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━  94/107 [huggingface-hub] 2025-09-09T14:11:54.5220590Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━━  94/107 [huggingface-hub] 2025-09-09T14:11:54.5221149Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━━  95/107 [aiohttp] 2025-09-09T14:11:54.5221700Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  97/107 [bitsandbytes] 2025-09-09T14:11:54.5222451Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  97/107 [bitsandbytes] 2025-09-09T14:11:54.5223012Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  97/107 [bitsandbytes] 2025-09-09T14:11:54.5223600Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  97/107 [bitsandbytes] 2025-09-09T14:11:54.5224178Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━━  97/107 [bitsandbytes] 2025-09-09T14:11:54.5224727Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━━  98/107 [accelerate] 2025-09-09T14:11:54.5225292Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5225849Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5226432Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5226991Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5227564Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5228123Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5228679Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5229241Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5229797Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5230352Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5231054Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5231608Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5232316Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5232887Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5233442Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5234011Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5234609Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5235191Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5235745Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5236317Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5236957Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5237508Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:11:54.5238065Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0650275Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0652581Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0653454Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0654097Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0655115Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0655701Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0656268Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━  99/107 [transformers] 2025-09-09T14:19:10.0656828Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺━━ 100/107 [datasets] 2025-09-09T14:19:10.0657486Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━━ 101/107 [DataProperty] 2025-09-09T14:19:10.0658309Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸━ 103/107 [peft] 2025-09-09T14:19:10.0658840Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╺ 105/107 [pytablewriter] 2025-09-09T14:19:10.0659367Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0659915Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0660442Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0660968Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0661462Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0661966Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0662464Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0662998Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0663501Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0664364Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0664880Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0665387Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0665888Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0666380Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0666880Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0667537Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0668046Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━╸ 106/107 [lm_eval] 2025-09-09T14:19:10.0668663Z  ━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━━ 107/107 [lm_eval] 2025-09-09T14:19:10.0669010Z [?25h 2025-09-09T14:19:10.0675803Z Successfully installed DataProperty-1.1.0 absl-py-2.3.1 accelerate-1.10.1 aiohappyeyeballs-2.6.1 aiohttp-3.12.15 aiosignal-1.4.0 async-timeout-5.0.1 attrs-25.3.0 bitsandbytes-0.47.0 blobfile-3.1.0 certifi-2025.8.3 cfgv-3.4.0 chardet-5.2.0 charset_normalizer-3.4.3 click-8.1.8 cmake-3.31.6 colorama-0.4.6 contourpy-1.3.0 cycler-0.12.1 datasets-3.6.0 dill-0.3.8 diskcache-5.6.3 distlib-0.4.0 evaluate-0.4.5 exceptiongroup-1.3.0 expecttest-0.3.0 fire-0.7.1 fonttools-4.59.2 frozenlist-1.7.0 fsspec-2025.3.0 hf-xet-1.1.9 huggingface-hub-0.34.4 hypothesis-6.138.15 identify-2.6.14 idna-3.10 importlib-resources-6.5.2 iniconfig-2.1.0 joblib-1.5.2 jsonlines-4.0.0 kiwisolver-1.4.7 lm_eval-0.4.9.1 lxml-6.0.1 matplotlib-3.9.4 mbstrdecoder-1.1.4 more_itertools-10.8.0 multidict-6.6.4 multiprocess-0.70.16 ninja-1.13.0 nltk-3.9.1 nodeenv-1.9.1 numexpr-2.10.2 numpy-2.0.2 packaging-25.0 pandas-2.3.2 parameterized-0.9.0 pathvalidate-3.3.1 peft-0.17.1 pillow-11.3.0 platformdirs-4.4.0 pluggy-1.6.0 portalocker-3.2.0 pre-commit-4.3.0 propcache-0.3.2 psutil-7.0.0 pyarrow-21.0.0 pybind11-3.0.1 pycocotools-2.0.10 pycryptodomex-3.23.0 pygments-2.19.2 pyparsing-3.2.3 pytablewriter-1.2.1 pytest-8.4.2 python-dateutil-2.9.0.post0 pytz-2025.2 pyyaml-6.0.2 regex-2025.9.1 requests-2.32.5 rouge-score-0.1.2 ruff-0.11.6 sacrebleu-2.5.1 safetensors-0.6.2 scikit-learn-1.6.1 scipy-1.13.1 sentencepiece-0.2.1 six-1.17.0 sortedcontainers-2.4.0 sqlitedict-2.1.0 tabledata-1.3.4 tabulate-0.9.0 tcolorpy-0.1.7 termcolor-3.1.0 threadpoolctl-3.6.0 tiktoken-0.11.0 tokenizers-0.22.0 tomli-2.2.1 tqdm-4.67.1 tqdm-multiprocess-0.0.11 transformers-4.56.1 typepy-1.3.4 tzdata-2025.2 unittest-xml-reporting-3.2.0 urllib3-2.5.0 virtualenv-20.34.0 word2number-1.1 xxhash-3.5.0 yarl-1.20.1 zstandard-0.24.0 2025-09-09T14:19:10.0683157Z WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. 2025-09-09T14:19:10.0684597Z + pip install . 2025-09-09T14:19:10.0684817Z Processing /pytorch/ao 2025-09-09T14:19:10.0685149Z Preparing metadata (setup.py) ... [?25l- done 2025-09-09T14:19:10.0685568Z [?25hBuilding wheels for collected packages: torchao 2025-09-09T14:19:10.0687579Z  DEPRECATION: Building 'torchao' using the legacy setup.py bdist_wheel mechanism, which will be removed in a future version. pip 25.3 will enforce this behaviour change. A possible replacement is to use the standardized build interface by setting the `--use-pep517` option, (possibly combined with `--no-build-isolation`), or adding a `pyproject.toml` file to the source tree of 'torchao'. Discussion can be found at https://github.com/pypa/pip/issues/6334 2025-09-09T14:19:10.0690957Z  Building wheel for torchao (setup.py) ... [?25l- \ | / - \ | / - \ | / - \ | / - \ | / - \ | done 2025-09-09T14:19:26.0177426Z [?25h Created wheel for torchao: filename=torchao-0.14.0+git7c05f81-cp39-abi3-linux_x86_64.whl size=7961001 sha256=a2c57d2972ef84f47cdbe29e8cbb870d82d4d54e42d4d805994dbc44a328ebc5 2025-09-09T14:19:26.0178575Z Stored in directory: /tmp/pip-ephem-wheel-cache-4rrsp0aw/wheels/4d/54/dc/0c70e60a8677bf126f1486798ebe76c8770ada66c7376b401d 2025-09-09T14:19:26.0179197Z Successfully built torchao 2025-09-09T14:19:26.0179476Z Installing collected packages: torchao 2025-09-09T14:19:26.0179834Z Successfully installed torchao-0.14.0+git7c05f81 2025-09-09T14:19:26.0181604Z WARNING: Running pip as the 'root' user can result in broken permissions and conflicting behaviour with the system package manager, possibly rendering your system unusable. It is recommended to use a virtual environment instead: https://pip.pypa.io/warnings/venv. Use the --root-user-action option if you know what you are doing and want to suppress this warning. 2025-09-09T14:19:26.0183065Z ++++ which conda 2025-09-09T14:19:26.0183307Z +++ dirname /opt/conda/condabin/conda 2025-09-09T14:19:26.0183611Z ++ dirname /opt/conda/condabin 2025-09-09T14:19:26.0183866Z + export CONDA=/opt/conda 2025-09-09T14:19:26.0184109Z + CONDA=/opt/conda 2025-09-09T14:19:26.0184595Z + export LD_LIBRARY_PATH=/opt/conda/lib/:/opt/rh/gcc-toolset-11/root/usr/lib64:/opt/rh/gcc-toolset-11/root/usr/lib: 2025-09-09T14:19:26.0185352Z + LD_LIBRARY_PATH=/opt/conda/lib/:/opt/rh/gcc-toolset-11/root/usr/lib64:/opt/rh/gcc-toolset-11/root/usr/lib: 2025-09-09T14:19:26.0185874Z + pytest test --verbose -s 2025-09-09T14:19:26.0186257Z ============================= test session starts ============================== 2025-09-09T14:19:26.0186785Z platform linux -- Python 3.9.23, pytest-8.4.2, pluggy-1.6.0 -- /opt/conda/envs/venv/bin/python3.9 2025-09-09T14:19:26.0187234Z cachedir: .pytest_cache 2025-09-09T14:19:26.0187811Z hypothesis profile 'ci' -> database=None, deadline=None, print_blob=True, derandomize=True, suppress_health_check=(HealthCheck.too_slow,) 2025-09-09T14:19:26.0188416Z rootdir: /pytorch/ao 2025-09-09T14:19:26.0188648Z plugins: hypothesis-6.138.15 2025-09-09T14:19:26.0188946Z collecting ...  2025-09-09T14:19:26.0189347Z collecting 0 items  2025-09-09T14:19:26.0189854Z collecting 26 items  2025-09-09T14:19:26.0190355Z collecting 26 items  2025-09-09T14:19:26.0190917Z collecting 273 items  2025-09-09T14:19:26.0191457Z collecting 682 items / 3 skipped  2025-09-09T14:19:26.0192022Z collecting 1031 items / 3 skipped  2025-09-09T14:19:26.0192597Z collecting 1072 items / 5 skipped  2025-09-09T14:19:26.0193843Z collecting 2986 items / 12 skipped NOTE: Using slow Hadamard transform for SpinQuant. For better performance on GPU, install `fast_hadamard_transform`: `pip install git+https://github.com/Dao-AILab/fast-hadamard-transform.git` 2025-09-09T14:19:26.0194797Z  2025-09-09T14:19:26.0195175Z collecting 4088 items / 12 skipped  2025-09-09T14:19:26.0195740Z collected 7126 items / 12 skipped  2025-09-09T14:19:26.0196505Z 2025-09-09T14:19:26.0197076Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config0] PASSED 2025-09-09T14:19:26.0200749Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config1] PASSED 2025-09-09T14:19:26.0201686Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config2] PASSED 2025-09-09T14:19:26.0202415Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config3] PASSED 2025-09-09T14:19:26.0203129Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config4] PASSED 2025-09-09T14:19:26.0203850Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config5] PASSED 2025-09-09T14:19:26.0204565Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config6] PASSED 2025-09-09T14:19:26.0205289Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config7] PASSED 2025-09-09T14:19:26.0206019Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config8] PASSED 2025-09-09T14:19:26.0206736Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config9] PASSED 2025-09-09T14:19:26.0207468Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config10] PASSED 2025-09-09T14:19:26.0208193Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config11] PASSED 2025-09-09T14:19:26.0208919Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config12] PASSED 2025-09-09T14:19:26.0209653Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config13] PASSED 2025-09-09T14:19:26.0210383Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config14] PASSED 2025-09-09T14:19:26.0211113Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config15] PASSED 2025-09-09T14:19:26.0211834Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config16] PASSED 2025-09-09T14:19:26.0212570Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config17] PASSED 2025-09-09T14:19:26.0213292Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config18] PASSED 2025-09-09T14:19:26.0214028Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config19] PASSED 2025-09-09T14:19:26.0214756Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config20] PASSED 2025-09-09T14:19:26.0215481Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config21] PASSED 2025-09-09T14:19:26.0216210Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config22] PASSED 2025-09-09T14:19:26.0216832Z test/core/test_config.py::test_disallowed_modules PASSED 2025-09-09T14:19:26.0217355Z test/core/test_config.py::test_version_mismatch PASSED 2025-09-09T14:19:26.0217878Z test/core/test_config.py::test_default_version PASSED 2025-09-09T14:19:26.0218623Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant0 PASSED 2025-09-09T14:19:26.0219569Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant1 PASSED 2025-09-09T14:19:26.0220508Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant2 PASSED 2025-09-09T14:19:26.0221440Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant3 PASSED 2025-09-09T14:19:26.0222374Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant4 PASSED 2025-09-09T14:19:26.0223303Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant5 PASSED 2025-09-09T14:19:26.0224236Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_copy__mismatch_metadata_apply_quant6 PASSED 2025-09-09T14:19:26.0225208Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_print_quantized_module PASSED 2025-09-09T14:19:26.0226042Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_register_new_dispatch PASSED 2025-09-09T14:19:26.0226975Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_tensor_core_layout_transpose PASSED 2025-09-09T14:19:26.0227849Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant0 PASSED 2025-09-09T14:19:26.0228858Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant1 PASSED 2025-09-09T14:19:26.0229738Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant2 PASSED 2025-09-09T14:19:26.0230617Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant3 PASSED 2025-09-09T14:19:26.0231502Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant4 PASSED 2025-09-09T14:19:26.0232375Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant5 PASSED 2025-09-09T14:19:26.0233258Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_test_copy__apply_apply_quant6 PASSED 2025-09-09T14:19:26.0234140Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_affine_quantized_intx_static PASSED 2025-09-09T14:19:26.0234998Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant0 PASSED 2025-09-09T14:19:26.0235826Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant1 PASSED 2025-09-09T14:19:26.0236888Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant2 PASSED 2025-09-09T14:19:26.0238219Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_to_device_apply_quant3 PASSED 2025-09-09T14:19:26.0239144Z test/dtypes/test_affine_quantized.py::TestAffineQuantized::test_weights_only PASSED 2025-09-09T14:19:26.0239978Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_alias_device_cpu_bfloat16 PASSED 2025-09-09T14:19:26.0240928Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_alias_device_cuda_bfloat16 PASSED 2025-09-09T14:19:26.0241886Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_flatten_unflatten_device_cpu_bfloat16 PASSED 2025-09-09T14:19:26.0242885Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_flatten_unflatten_device_cuda_bfloat16 PASSED 2025-09-09T14:19:26.0243830Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_matmul_device_cuda_bfloat16 PASSED 2025-09-09T14:19:26.0764652Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_mm_int4wo_device_cuda_bfloat16 PASSED 2025-09-09T14:19:26.0765790Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_and_copy_int4wo_device_cuda_bfloat16 PASSED 2025-09-09T14:19:26.0766784Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_gemlite_device_cuda_bfloat16 SKIPPED 2025-09-09T14:19:26.0767908Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_gemlite_device_cuda_float16 SKIPPED 2025-09-09T14:19:26.0768903Z test/dtypes/test_affine_quantized.py::TestAffineQuantizedBasic::test_slice_int4wo_device_cuda_bfloat16 PASSED 2025-09-09T14:19:26.0770123Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e4m3fn_bfloat16 SKIPPED 2025-09-09T14:19:26.0771403Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e4m3fn_float32 SKIPPED 2025-09-09T14:19:26.0772836Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e5m2_bfloat16 SKIPPED 2025-09-09T14:19:26.0774025Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_choose_scale_float8_bounds_float8_e5m2_float32 SKIPPED 2025-09-09T14:19:26.0775667Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size0 SKIPPED 2025-09-09T14:19:26.0777097Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size1 SKIPPED 2025-09-09T14:19:26.0778411Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size2 SKIPPED 2025-09-09T14:19:26.0779714Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_bfloat16_block_size3 SKIPPED 2025-09-09T14:19:26.0781009Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size0 SKIPPED 2025-09-09T14:19:26.0782552Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size1 SKIPPED 2025-09-09T14:19:26.0784185Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size2 SKIPPED 2025-09-09T14:19:26.0785472Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e4m3fn_float32_block_size3 SKIPPED 2025-09-09T14:19:26.0786754Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size0 SKIPPED 2025-09-09T14:19:26.0788039Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size1 SKIPPED 2025-09-09T14:19:26.0789314Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size2 SKIPPED 2025-09-09T14:19:26.0790764Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_bfloat16_block_size3 SKIPPED 2025-09-09T14:19:26.0792055Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size0 SKIPPED 2025-09-09T14:19:26.0793451Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size1 SKIPPED 2025-09-09T14:19:26.0794739Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size2 SKIPPED 2025-09-09T14:19:26.0796009Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_float8_e5m2_float32_block_size3 SKIPPED 2025-09-09T14:19:26.0797866Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_dequantize_affine_float8_scale_broadcasting SKIPPED 2025-09-09T14:19:26.0799208Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity0_float8_config_version_1 SKIPPED 2025-09-09T14:19:26.0800652Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity0_float8_config_version_2 SKIPPED 2025-09-09T14:19:26.0801995Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity1_float8_config_version_1 SKIPPED 2025-09-09T14:19:26.0803428Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_expected_kernels_on_gpu_granularity1_float8_config_version_2 SKIPPED 2025-09-09T14:19:26.0804647Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_basic_granularity0 SKIPPED 2025-09-09T14:19:26.0806058Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_basic_granularity1 SKIPPED 2025-09-09T14:19:26.0807174Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_edge_cases SKIPPED 2025-09-09T14:19:26.0808510Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_functional_correctness_granularity0 SKIPPED 2025-09-09T14:19:26.0809805Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_functional_correctness_granularity1 SKIPPED 2025-09-09T14:19:26.0810989Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_per_row SKIPPED 2025-09-09T14:19:26.0812138Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_float8_tensor_slicing_per_tensor SKIPPED 2025-09-09T14:19:26.0813470Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.0815002Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.0816417Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.0817831Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.0819291Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.0820815Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.0822326Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.0823735Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.0825142Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.0826658Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.0828076Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.0829589Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.0831035Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.0832445Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.0834085Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1113846Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_static_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1115661Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1117755Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1119442Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1120954Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1122423Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1124002Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1125570Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1127024Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1128447Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1129930Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1131385Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1132894Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1134310Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1135718Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1137231Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1138640Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1140145Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1141538Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1143107Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1144576Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1146097Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1147498Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1149003Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1150403Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_static_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1151883Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1153441Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1154911Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1156650Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1167343Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_sizes0 SKIPPED 2025-09-09T14:19:26.1169043Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_sizes1 SKIPPED 2025-09-09T14:19:26.1170606Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_sizes0 SKIPPED 2025-09-09T14:19:26.1172051Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_sizes1 SKIPPED 2025-09-09T14:19:26.1173276Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_fp8_weight_dimension_warning SKIPPED 2025-09-09T14:19:26.1174321Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_invalid_granularity SKIPPED 2025-09-09T14:19:26.1175465Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mismatched_granularity SKIPPED 2025-09-09T14:19:26.1176701Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape0_bias_False SKIPPED 2025-09-09T14:19:26.1178238Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape0_bias_True SKIPPED 2025-09-09T14:19:26.1179647Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape1_bias_False SKIPPED 2025-09-09T14:19:26.1181325Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape1_bias_True SKIPPED 2025-09-09T14:19:26.1182998Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape2_bias_False SKIPPED 2025-09-09T14:19:26.1184419Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape2_bias_True SKIPPED 2025-09-09T14:19:26.1185947Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape3_bias_False SKIPPED 2025-09-09T14:19:26.1187357Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape3_bias_True SKIPPED 2025-09-09T14:19:26.1188763Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape4_bias_False SKIPPED 2025-09-09T14:19:26.1190307Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_1024_out_features_512_leading_shape4_bias_True SKIPPED 2025-09-09T14:19:26.1191777Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape0_bias_False SKIPPED 2025-09-09T14:19:52.5201137Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape0_bias_True SKIPPED 2025-09-09T14:19:52.5203700Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape1_bias_False SKIPPED 2025-09-09T14:19:52.5205164Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape1_bias_True SKIPPED 2025-09-09T14:19:52.5206577Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape2_bias_False SKIPPED 2025-09-09T14:19:52.5207981Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape2_bias_True SKIPPED 2025-09-09T14:19:52.5209368Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape3_bias_False SKIPPED 2025-09-09T14:19:52.5210752Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape3_bias_True SKIPPED 2025-09-09T14:19:52.5212141Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape4_bias_False SKIPPED 2025-09-09T14:19:52.5213526Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_256_out_features_768_leading_shape4_bias_True SKIPPED 2025-09-09T14:19:52.5214923Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape0_bias_False SKIPPED 2025-09-09T14:19:52.5216315Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape0_bias_True SKIPPED 2025-09-09T14:19:52.5217756Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape1_bias_False SKIPPED 2025-09-09T14:19:52.5219553Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape1_bias_True SKIPPED 2025-09-09T14:19:52.5221117Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape2_bias_False SKIPPED 2025-09-09T14:19:52.5222512Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape2_bias_True SKIPPED 2025-09-09T14:19:52.5223910Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape3_bias_False SKIPPED 2025-09-09T14:19:52.5225320Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape3_bias_True SKIPPED 2025-09-09T14:19:52.5226713Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape4_bias_False SKIPPED 2025-09-09T14:19:52.5228111Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_mm_float8dq_per_row_in_features_512_out_features_1024_leading_shape4_bias_True SKIPPED 2025-09-09T14:19:52.5229292Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_per_row_with_float32 SKIPPED 2025-09-09T14:19:52.5230310Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_preprocess_scale_3d_reshape PASSED 2025-09-09T14:19:52.5231354Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_serialization_mode_dynamic SKIPPED 2025-09-09T14:19:52.5232389Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_serialization_mode_static SKIPPED 2025-09-09T14:19:52.5233458Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_serialization_mode_weight-only SKIPPED 2025-09-09T14:19:52.5234518Z test/dtypes/test_affine_quantized_float.py::TestAffineQuantizedFloat8Compile::test_unsupported_granularity SKIPPED 2025-09-09T14:19:52.5235847Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8woAffineQuantizedTensorParallel::test_tp_bfloat16 I0909 14:19:26.182420 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 1119 2025-09-09T14:19:52.5237259Z I0909 14:19:26.210572 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 1120 2025-09-09T14:19:52.5238089Z I0909 14:19:26.239774 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 2 with pid 1121 2025-09-09T14:19:52.5238902Z I0909 14:19:26.269310 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 3 with pid 1122 2025-09-09T14:19:52.5241516Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:19:52.5243847Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:19:52.5247222Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:19:52.5249624Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:19:52.5251939Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:19:52.5254250Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:19:52.5256579Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:19:52.5258887Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:19:52.5259218Z PASSED 2025-09-09T14:19:52.5260161Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8woAffineQuantizedTensorParallel::test_tp_float16 I0909 14:19:48.236143 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 3569 2025-09-09T14:19:52.5261432Z I0909 14:19:48.261891 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 3570 2025-09-09T14:19:52.5262259Z I0909 14:19:48.289918 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 2 with pid 3571 2025-09-09T14:19:52.5263081Z I0909 14:19:48.319525 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 3 with pid 3572 2025-09-09T14:20:34.7915625Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7918136Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7920481Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7922785Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7925718Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7928031Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7930348Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7932653Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7933167Z PASSED 2025-09-09T14:20:34.7934131Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8woAffineQuantizedTensorParallel::test_tp_float32 I0909 14:20:08.885329 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 5877 2025-09-09T14:20:34.7935452Z I0909 14:20:08.912086 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 5878 2025-09-09T14:20:34.7936282Z I0909 14:20:08.941798 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 2 with pid 5879 2025-09-09T14:20:34.7937111Z I0909 14:20:08.971022 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 3 with pid 5880 2025-09-09T14:20:34.7939689Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7941989Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7944305Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7946612Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7948904Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7951288Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7953667Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7955965Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7956366Z PASSED 2025-09-09T14:20:34.7957065Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt4woAffineQuantizedTensorParallel::test_tp_bfloat16 SKIPPED 2025-09-09T14:20:34.7958139Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestGemliteLayoutTensorParallel::test_tp_gemlite_float16 SKIPPED 2025-09-09T14:20:34.7959490Z test/dtypes/test_affine_quantized_tensor_parallel.py::TestInt8dqAffineQuantizedTensorParallel::test_tp_bfloat16 I0909 14:20:30.436056 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 8223 2025-09-09T14:20:34.7960763Z I0909 14:20:30.462837 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 8224 2025-09-09T14:20:34.7961583Z I0909 14:20:30.491956 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 2 with pid 8225 2025-09-09T14:20:34.7962401Z I0909 14:20:30.520670 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 3 with pid 8226 2025-09-09T14:20:34.7965165Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7967461Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:20:34.7969765Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:20:34.7972066Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:22:05.7748480Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:22:05.7751961Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:22:05.7755098Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T14:22:05.7758125Z _C._set_float32_matmul_precision(precision) 2025-09-09T14:22:05.7758724Z PASSED 2025-09-09T14:22:05.7759183Z test/dtypes/test_bitpacking.py::test_CPU[0-1] PASSED 2025-09-09T14:22:05.7759802Z test/dtypes/test_bitpacking.py::test_CPU[0-2] PASSED 2025-09-09T14:22:05.7760415Z test/dtypes/test_bitpacking.py::test_CPU[0-3] PASSED 2025-09-09T14:22:05.7761021Z test/dtypes/test_bitpacking.py::test_CPU[0-4] PASSED 2025-09-09T14:22:05.7761623Z test/dtypes/test_bitpacking.py::test_CPU[0-5] PASSED 2025-09-09T14:22:05.7762238Z test/dtypes/test_bitpacking.py::test_CPU[0-6] PASSED 2025-09-09T14:22:05.7762843Z test/dtypes/test_bitpacking.py::test_CPU[0-7] PASSED 2025-09-09T14:22:05.7763448Z test/dtypes/test_bitpacking.py::test_CPU[-1-1] PASSED 2025-09-09T14:22:05.7764253Z test/dtypes/test_bitpacking.py::test_CPU[-1-2] PASSED 2025-09-09T14:22:05.7764868Z test/dtypes/test_bitpacking.py::test_CPU[-1-3] PASSED 2025-09-09T14:22:05.7765481Z test/dtypes/test_bitpacking.py::test_CPU[-1-4] PASSED 2025-09-09T14:22:05.7766088Z test/dtypes/test_bitpacking.py::test_CPU[-1-5] PASSED 2025-09-09T14:22:05.7766712Z test/dtypes/test_bitpacking.py::test_CPU[-1-6] PASSED 2025-09-09T14:22:05.7767338Z test/dtypes/test_bitpacking.py::test_CPU[-1-7] PASSED 2025-09-09T14:22:05.7767941Z test/dtypes/test_bitpacking.py::test_CPU[1-1] PASSED 2025-09-09T14:22:05.7768549Z test/dtypes/test_bitpacking.py::test_CPU[1-2] PASSED 2025-09-09T14:22:05.7769151Z test/dtypes/test_bitpacking.py::test_CPU[1-3] PASSED 2025-09-09T14:22:05.7769764Z test/dtypes/test_bitpacking.py::test_CPU[1-4] PASSED 2025-09-09T14:22:05.7770365Z test/dtypes/test_bitpacking.py::test_CPU[1-5] PASSED 2025-09-09T14:22:05.7770978Z test/dtypes/test_bitpacking.py::test_CPU[1-6] PASSED 2025-09-09T14:22:05.7771578Z test/dtypes/test_bitpacking.py::test_CPU[1-7] PASSED 2025-09-09T14:22:05.7772188Z test/dtypes/test_bitpacking.py::test_GPU[0-1] PASSED 2025-09-09T14:22:05.7772798Z test/dtypes/test_bitpacking.py::test_GPU[0-2] PASSED 2025-09-09T14:22:05.7773395Z test/dtypes/test_bitpacking.py::test_GPU[0-3] PASSED 2025-09-09T14:22:05.7774024Z test/dtypes/test_bitpacking.py::test_GPU[0-4] PASSED 2025-09-09T14:22:05.7774624Z test/dtypes/test_bitpacking.py::test_GPU[0-5] PASSED 2025-09-09T14:22:05.7775232Z test/dtypes/test_bitpacking.py::test_GPU[0-6] PASSED 2025-09-09T14:22:05.7775834Z test/dtypes/test_bitpacking.py::test_GPU[0-7] PASSED 2025-09-09T14:22:05.7776453Z test/dtypes/test_bitpacking.py::test_GPU[-1-1] PASSED 2025-09-09T14:22:05.7777061Z test/dtypes/test_bitpacking.py::test_GPU[-1-2] PASSED 2025-09-09T14:22:05.7777672Z test/dtypes/test_bitpacking.py::test_GPU[-1-3] PASSED 2025-09-09T14:22:05.7778295Z test/dtypes/test_bitpacking.py::test_GPU[-1-4] PASSED 2025-09-09T14:22:05.7778895Z test/dtypes/test_bitpacking.py::test_GPU[-1-5] PASSED 2025-09-09T14:22:05.7779512Z test/dtypes/test_bitpacking.py::test_GPU[-1-6] PASSED 2025-09-09T14:22:05.7780111Z test/dtypes/test_bitpacking.py::test_GPU[-1-7] PASSED 2025-09-09T14:22:05.7780723Z test/dtypes/test_bitpacking.py::test_GPU[1-1] PASSED 2025-09-09T14:22:05.7781481Z test/dtypes/test_bitpacking.py::test_GPU[1-2] PASSED 2025-09-09T14:22:05.7782085Z test/dtypes/test_bitpacking.py::test_GPU[1-3] PASSED 2025-09-09T14:22:05.7782687Z test/dtypes/test_bitpacking.py::test_GPU[1-4] PASSED 2025-09-09T14:22:05.7783440Z test/dtypes/test_bitpacking.py::test_GPU[1-5] PASSED 2025-09-09T14:22:05.7784058Z test/dtypes/test_bitpacking.py::test_GPU[1-6] PASSED 2025-09-09T14:22:05.7784658Z test/dtypes/test_bitpacking.py::test_GPU[1-7] PASSED 2025-09-09T14:22:05.7785296Z test/dtypes/test_bitpacking.py::test_compile[0-1] PASSED 2025-09-09T14:22:05.7785948Z test/dtypes/test_bitpacking.py::test_compile[0-2] PASSED 2025-09-09T14:22:05.7786608Z test/dtypes/test_bitpacking.py::test_compile[0-3] PASSED 2025-09-09T14:22:05.7787261Z test/dtypes/test_bitpacking.py::test_compile[0-4] PASSED 2025-09-09T14:22:05.7787916Z test/dtypes/test_bitpacking.py::test_compile[0-5] PASSED 2025-09-09T14:22:05.7788572Z test/dtypes/test_bitpacking.py::test_compile[0-6] PASSED 2025-09-09T14:22:05.7789215Z test/dtypes/test_bitpacking.py::test_compile[0-7] PASSED 2025-09-09T14:22:05.7789879Z test/dtypes/test_bitpacking.py::test_compile[-1-1] PASSED 2025-09-09T14:22:05.7790542Z test/dtypes/test_bitpacking.py::test_compile[-1-2] PASSED 2025-09-09T14:22:05.7791203Z test/dtypes/test_bitpacking.py::test_compile[-1-3] PASSED 2025-09-09T14:22:05.7791863Z test/dtypes/test_bitpacking.py::test_compile[-1-4] PASSED 2025-09-09T14:22:05.7792528Z test/dtypes/test_bitpacking.py::test_compile[-1-5] PASSED 2025-09-09T14:22:05.7793193Z test/dtypes/test_bitpacking.py::test_compile[-1-6] PASSED 2025-09-09T14:22:05.7793843Z test/dtypes/test_bitpacking.py::test_compile[-1-7] PASSED 2025-09-09T14:22:05.7794499Z test/dtypes/test_bitpacking.py::test_compile[1-1] PASSED 2025-09-09T14:22:05.7795154Z test/dtypes/test_bitpacking.py::test_compile[1-2] PASSED 2025-09-09T14:22:05.7795810Z test/dtypes/test_bitpacking.py::test_compile[1-3] PASSED 2025-09-09T14:22:05.7796527Z test/dtypes/test_bitpacking.py::test_compile[1-4] PASSED 2025-09-09T14:22:05.7797177Z test/dtypes/test_bitpacking.py::test_compile[1-5] PASSED 2025-09-09T14:22:05.7797842Z test/dtypes/test_bitpacking.py::test_compile[1-6] PASSED 2025-09-09T14:22:05.7798488Z test/dtypes/test_bitpacking.py::test_compile[1-7] PASSED 2025-09-09T14:22:05.7799448Z test/dtypes/test_bitpacking.py::test_pack_example tensor([ 0, 105, 151, 37], device='cuda:0', dtype=torch.uint8) tensor([ 39, 146], device='cuda:0', dtype=torch.uint8) 2025-09-09T14:22:05.7800304Z PASSED 2025-09-09T14:22:05.7800941Z test/dtypes/test_bitpacking.py::test_pack_example_CPU tensor([ 0, 105, 151, 37], dtype=torch.uint8) tensor([ 39, 146], dtype=torch.uint8) 2025-09-09T14:22:05.7801692Z PASSED 2025-09-09T14:22:05.7802552Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_False_bfloat16 PASSED 2025-09-09T14:22:05.7803944Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_False_float16 PASSED 2025-09-09T14:22:05.7805327Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_True_bfloat16 PASSED 2025-09-09T14:22:05.7806711Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_2_mbits_2_bias_True_float16 PASSED 2025-09-09T14:22:05.7808099Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_False_bfloat16 PASSED 2025-09-09T14:22:05.7809483Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_False_float16 PASSED 2025-09-09T14:22:05.7810870Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_True_bfloat16 PASSED 2025-09-09T14:22:05.7812334Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_fpx_weight_only_ebits_3_mbits_2_bias_True_float16 PASSED 2025-09-09T14:22:05.7813894Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cpu PASSED 2025-09-09T14:22:05.7815364Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cuda PASSED 2025-09-09T14:22:05.7816820Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cpu PASSED 2025-09-09T14:22:05.7818283Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cuda PASSED 2025-09-09T14:22:05.7819729Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_2_mbits_2_device_cpu PASSED 2025-09-09T14:22:05.7821168Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_2_mbits_2_device_cuda PASSED 2025-09-09T14:22:05.7822616Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_3_mbits_2_device_cpu PASSED 2025-09-09T14:22:05.7824106Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_from_tc_floatx_correctness_ebits_3_mbits_2_device_cuda PASSED 2025-09-09T14:22:05.7825443Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_pack_tc_fp6_correctness_device_cpu PASSED 2025-09-09T14:22:05.7826692Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_pack_tc_fp6_correctness_device_cuda PASSED 2025-09-09T14:23:52.3022133Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_copy_device_ebits_2_mbits_2 PASSED 2025-09-09T14:23:52.3023433Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_copy_device_ebits_3_mbits_2 PASSED 2025-09-09T14:23:52.3024840Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cpu PASSED 2025-09-09T14:23:52.3026351Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_2_mbits_2_device_cuda PASSED 2025-09-09T14:23:52.3027654Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cpu PASSED 2025-09-09T14:23:52.3028766Z test/dtypes/test_floatx.py::TestFloatxTensorCoreAQTTensorImpl::test_to_scaled_tc_floatx_compile_ebits_3_mbits_2_device_cuda PASSED 2025-09-09T14:23:52.3029662Z test/dtypes/test_nf4.py::TestNF4Linear::test_backward_dtype_match_bfloat16 PASSED 2025-09-09T14:23:52.3030357Z test/dtypes/test_nf4.py::TestNF4Linear::test_backward_dtype_match_float16 PASSED 2025-09-09T14:23:52.3031049Z test/dtypes/test_nf4.py::TestNF4Linear::test_backward_dtype_match_float32 PASSED 2025-09-09T14:23:52.3031834Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape0_chunk_size_16 PASSED 2025-09-09T14:23:52.3032700Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape0_chunk_size_32 PASSED 2025-09-09T14:23:52.3033560Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape0_chunk_size_8 PASSED 2025-09-09T14:23:52.3034419Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape1_chunk_size_16 PASSED 2025-09-09T14:23:52.3035274Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape1_chunk_size_32 PASSED 2025-09-09T14:23:52.3036191Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_bfloat16_shape1_chunk_size_8 PASSED 2025-09-09T14:23:52.3037356Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape0_chunk_size_16 PASSED 2025-09-09T14:23:52.3038210Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape0_chunk_size_32 PASSED 2025-09-09T14:23:52.3039224Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape0_chunk_size_8 PASSED 2025-09-09T14:23:52.3040076Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape1_chunk_size_16 PASSED 2025-09-09T14:23:52.3040927Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape1_chunk_size_32 PASSED 2025-09-09T14:23:52.3042008Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float16_shape1_chunk_size_8 PASSED 2025-09-09T14:23:52.3043031Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape0_chunk_size_16 PASSED 2025-09-09T14:23:52.3044069Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape0_chunk_size_32 PASSED 2025-09-09T14:23:52.3045108Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape0_chunk_size_8 PASSED 2025-09-09T14:23:52.3046144Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape1_chunk_size_16 PASSED 2025-09-09T14:23:52.3047175Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape1_chunk_size_32 PASSED 2025-09-09T14:23:52.3048199Z test/dtypes/test_nf4.py::TestNF4Linear::test_chunk_size_equivalence_float32_shape1_chunk_size_8 PASSED 2025-09-09T14:23:52.3049091Z test/dtypes/test_nf4.py::TestNF4Linear::test_empty_like_input_size0 PASSED 2025-09-09T14:23:52.3049722Z test/dtypes/test_nf4.py::TestNF4Linear::test_empty_like_input_size1 PASSED 2025-09-09T14:23:52.3050401Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_diff_meta_bfloat16 PASSED 2025-09-09T14:23:52.3051135Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_diff_meta_float16 PASSED 2025-09-09T14:23:52.3051862Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_diff_meta_float32 PASSED 2025-09-09T14:23:52.3052567Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_same_meta_bfloat16 PASSED 2025-09-09T14:23:52.3053269Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_same_meta_float16 PASSED 2025-09-09T14:23:52.3053967Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_nf4_same_meta_float32 PASSED 2025-09-09T14:23:52.3054662Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_state_dicts_bfloat16 PASSED 2025-09-09T14:23:52.3055349Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_state_dicts_float16 PASSED 2025-09-09T14:23:52.3056038Z test/dtypes/test_nf4.py::TestNF4Linear::test_load_from_state_dicts_float32 PASSED 2025-09-09T14:23:52.3056708Z test/dtypes/test_nf4.py::TestNF4Linear::test_nf4_bnb_linear_bfloat16 SKIPPED 2025-09-09T14:23:52.3057372Z test/dtypes/test_nf4.py::TestNF4Linear::test_nf4_bnb_linear_float16 SKIPPED 2025-09-09T14:23:52.3058022Z test/dtypes/test_nf4.py::TestNF4Linear::test_nf4_bnb_linear_float32 SKIPPED 2025-09-09T14:23:52.3058681Z test/dtypes/test_nf4.py::TestNF4Linear::test_output_dtype_match_bfloat16 PASSED 2025-09-09T14:23:52.3059349Z test/dtypes/test_nf4.py::TestNF4Linear::test_output_dtype_match_float16 PASSED 2025-09-09T14:23:52.3060009Z test/dtypes/test_nf4.py::TestNF4Linear::test_output_dtype_match_float32 PASSED 2025-09-09T14:23:52.3060677Z test/dtypes/test_nf4.py::TestNF4Linear::test_quantize_api_compile_False PASSED 2025-09-09T14:23:52.3061334Z test/dtypes/test_nf4.py::TestNF4Linear::test_quantize_api_compile_True PASSED 2025-09-09T14:23:52.3062055Z test/dtypes/test_nf4.py::TestNF4Linear::test_reconstruction_qlora_vs_bnb_bfloat16 SKIPPED 2025-09-09T14:23:52.3062917Z test/dtypes/test_nf4.py::TestNF4Linear::test_reconstruction_qlora_vs_bnb_float16 SKIPPED 2025-09-09T14:23:52.3063670Z test/dtypes/test_nf4.py::TestNF4Linear::test_reconstruction_qlora_vs_bnb_float32 SKIPPED 2025-09-09T14:23:52.3064592Z test/dtypes/test_nf4.py::TestNF4Linear::test_register_nf4_as_param_bfloat16 PASSED 2025-09-09T14:23:52.3065411Z test/dtypes/test_nf4.py::TestNF4Linear::test_register_nf4_as_param_float16 PASSED 2025-09-09T14:23:52.3066107Z test/dtypes/test_nf4.py::TestNF4Linear::test_register_nf4_as_param_float32 PASSED 2025-09-09T14:23:52.3066785Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_bfloat16 PASSED 2025-09-09T14:23:52.3067430Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_compile_bfloat16 Autotune Choices Stats: 2025-09-09T14:23:52.3068630Z {"num_choices": 13, "num_triton_choices": 12, "best_kernel": "triton_mm_0", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:23:52.3069615Z AUTOTUNE mm(64x32, 32x32) 2025-09-09T14:23:52.3069844Z strides: [32, 1], [1, 32] 2025-09-09T14:23:52.3070099Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:23:52.3070738Z triton_mm_0 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:23:52.3071708Z triton_mm_1 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:23:52.3072668Z triton_mm_3 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:23:52.3073612Z triton_mm_4 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:23:52.3074569Z triton_mm_5 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:23:52.3075528Z triton_mm_6 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:23:52.3076522Z triton_mm_7 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:23:52.3077473Z triton_mm_9 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:23:52.3078426Z triton_mm_10 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:23:52.3079384Z triton_mm_11 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:23:52.3080234Z SingleProcess AUTOTUNE benchmarking takes 0.1631 seconds and 0.5574 seconds precompiling for 13 choices 2025-09-09T14:23:52.3080765Z PASSED 2025-09-09T14:23:52.3081254Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_compile_float16 Autotune Choices Stats: 2025-09-09T14:24:01.4715253Z {"num_choices": 13, "num_triton_choices": 12, "best_kernel": "triton_mm_20", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:24:01.4717142Z AUTOTUNE mm(64x32, 32x32) 2025-09-09T14:24:01.4717428Z strides: [32, 1], [1, 32] 2025-09-09T14:24:01.4717770Z dtypes: torch.float16, torch.float16 2025-09-09T14:24:01.4718739Z triton_mm_20 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:24:01.4720265Z triton_mm_21 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:24:01.4721510Z triton_mm_22 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:24:01.4722725Z triton_mm_23 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:24:01.4723934Z triton_mm_12 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:24:01.4725142Z triton_mm_13 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:24:01.4726349Z triton_mm_14 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:24:01.4727624Z triton_mm_15 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:24:01.4728846Z triton_mm_16 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:24:01.4730074Z triton_mm_17 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:24:01.4731149Z SingleProcess AUTOTUNE benchmarking takes 0.2903 seconds and 0.3459 seconds precompiling for 13 choices 2025-09-09T14:24:01.4732062Z PASSED 2025-09-09T14:24:01.4732627Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_compile_float32 Autotune Choices Stats: 2025-09-09T14:24:01.4734153Z {"num_choices": 13, "num_triton_choices": 12, "best_kernel": "triton_mm_30", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.02454400062561035, "best_triton_pos": 0} 2025-09-09T14:24:01.4735443Z AUTOTUNE mm(64x32, 32x32) 2025-09-09T14:24:01.4735723Z strides: [32, 1], [1, 32] 2025-09-09T14:24:01.4736141Z dtypes: torch.float32, torch.float32 2025-09-09T14:24:01.4737404Z triton_mm_30 0.0245 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:24:01.4738808Z triton_mm_24 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:24:01.4740148Z triton_mm_25 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:24:01.4741468Z triton_mm_26 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:24:01.4742769Z triton_mm_27 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:24:01.4744171Z triton_mm_28 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:24:01.4754245Z triton_mm_29 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:24:01.4755548Z triton_mm_31 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:24:01.4756882Z triton_mm_32 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:24:01.4758116Z triton_mm_33 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:24:01.4759211Z SingleProcess AUTOTUNE benchmarking takes 0.1609 seconds and 0.2938 seconds precompiling for 13 choices 2025-09-09T14:24:01.4759945Z PASSED 2025-09-09T14:24:01.4760516Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_float16 PASSED 2025-09-09T14:24:01.4761346Z test/dtypes/test_nf4.py::TestNF4Linear::test_smoketest_linear_float32 PASSED 2025-09-09T14:24:01.4762135Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_bfloat16 PASSED 2025-09-09T14:24:01.4762874Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_device PASSED 2025-09-09T14:24:01.4763609Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float16 PASSED 2025-09-09T14:24:01.4764579Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float32 PASSED 2025-09-09T14:24:01.4765330Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_dtype_bfloat16 PASSED 2025-09-09T14:24:01.4766092Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_dtype_float16 PASSED 2025-09-09T14:24:01.4766878Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_dtype_float32 PASSED 2025-09-09T14:24:01.4767622Z test/dtypes/test_nf4.py::TestFSDPOps::test_pin_memory PASSED 2025-09-09T14:24:01.4768417Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_2d_view_valid_input_size0 PASSED 2025-09-09T14:24:01.4769334Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_invalid_input_size0 PASSED 2025-09-09T14:24:01.4770274Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_invalid_input_size1 PASSED 2025-09-09T14:24:01.4771189Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_valid_input_size1 PASSED 2025-09-09T14:24:01.4772100Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_valid_input_size2 PASSED 2025-09-09T14:24:01.4773040Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_as_strided_valid_input_size_262144 PASSED 2025-09-09T14:24:01.4773958Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_deepcopy_input_size1 PASSED 2025-09-09T14:24:01.4774799Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_deepcopy_input_size2 PASSED 2025-09-09T14:24:01.4775660Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_deepcopy_input_size_262144 PASSED 2025-09-09T14:24:01.4776577Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_invalid_input_size1 PASSED 2025-09-09T14:24:01.4777493Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_invalid_input_size2 PASSED 2025-09-09T14:24:01.4778452Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_invalid_input_size_262144 PASSED 2025-09-09T14:24:01.4779391Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_valid_input_size1 PASSED 2025-09-09T14:24:01.4780282Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_valid_input_size2 PASSED 2025-09-09T14:24:01.4781216Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_new_zeros_valid_input_size_262144 PASSED 2025-09-09T14:24:01.4782352Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_1d_invalid PASSED 2025-09-09T14:24:01.4783146Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_2d_invalid PASSED 2025-09-09T14:24:01.4784090Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_valid_input_size1 PASSED 2025-09-09T14:24:01.4784941Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_valid_input_size2 PASSED 2025-09-09T14:24:01.4785822Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_slice_valid_input_size_262144 PASSED 2025-09-09T14:24:01.4786701Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_view_invalid_input_size0 PASSED 2025-09-09T14:24:01.4787564Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_view_valid_input_size0 PASSED 2025-09-09T14:24:01.4788409Z test/dtypes/test_nf4.py::TestFSDPOps::test_tensor_view_valid_input_size1 PASSED 2025-09-09T14:24:01.4789169Z test/dtypes/test_nf4.py::TestFSDPOps::test_to_cpu PASSED 2025-09-09T14:24:01.4789827Z test/dtypes/test_nf4.py::TestFSDPOps::test_to_cuda PASSED 2025-09-09T14:24:01.4790485Z test/dtypes/test_nf4.py::TestFSDPOps::test_to_module PASSED 2025-09-09T14:24:01.4791291Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_3d_input_size0 PASSED 2025-09-09T14:24:01.4792271Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_divide_input_size1 PASSED 2025-09-09T14:24:01.4793220Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_divide_input_size2 PASSED 2025-09-09T14:29:02.4148493Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_invalid_divide_input_size_261632 PASSED 2025-09-09T14:29:02.4149367Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_valid_input_size1 PASSED 2025-09-09T14:29:02.4150133Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_valid_input_size2 PASSED 2025-09-09T14:29:02.4150936Z test/dtypes/test_nf4.py::TestFSDPOps::test_torch_chunk_valid_input_size_262144 PASSED 2025-09-09T14:29:02.4151943Z test/dtypes/test_nf4.py::TestQLoRA::test_qlora_fsdp2 I0909 14:24:01.520117 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 14577 2025-09-09T14:29:02.4153115Z I0909 14:24:01.556196 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 14578 2025-09-09T14:29:02.4153675Z dist init r=0, world=2 2025-09-09T14:29:02.4153963Z dist init r=1, world=2 2025-09-09T14:29:02.4154899Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:29:02.4155889Z warnings.warn( # warn only once 2025-09-09T14:29:02.4156908Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:29:02.4157902Z warnings.warn( # warn only once 2025-09-09T14:29:02.4158290Z PASSED 2025-09-09T14:29:02.4159009Z test/dtypes/test_nf4.py::TestComm::test_comm I0909 14:24:11.507239 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 14758 2025-09-09T14:29:02.4160134Z I0909 14:24:11.543209 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 14759 2025-09-09T14:29:02.4160752Z dist init r=0, world=2 2025-09-09T14:29:02.4160979Z dist init r=1, world=2 2025-09-09T14:29:02.4161894Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:29:02.4162885Z warnings.warn( # warn only once 2025-09-09T14:29:02.4164135Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:29:02.4165570Z warnings.warn( # warn only once 2025-09-09T14:29:02.4165951Z PASSED 2025-09-09T14:29:02.4166646Z test/dtypes/test_uint4.py::TestUInt4::test_basic_tensor_ops SKIPPED 2025-09-09T14:29:02.4167334Z test/dtypes/test_uint4.py::TestUInt4::test_gpu_quant SKIPPED (FAILED...) 2025-09-09T14:29:02.4168028Z test/dtypes/test_uint4.py::TestUInt4::test_pt2e_quant SKIPPED (FAILE...) 2025-09-09T14:29:02.4168786Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype0] PASSED 2025-09-09T14:29:02.4169593Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype1] PASSED 2025-09-09T14:29:02.4170385Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype2] PASSED 2025-09-09T14:29:02.4171218Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype3] PASSED 2025-09-09T14:29:02.4172020Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype4] PASSED 2025-09-09T14:29:02.4172811Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype5] PASSED 2025-09-09T14:29:02.4173620Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[32-dtype6] PASSED 2025-09-09T14:29:02.4174421Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype0] PASSED 2025-09-09T14:29:02.4175223Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype1] PASSED 2025-09-09T14:29:02.4176020Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype2] PASSED 2025-09-09T14:29:02.4176795Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype3] PASSED 2025-09-09T14:29:02.4177574Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype4] PASSED 2025-09-09T14:29:02.4178358Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype5] PASSED 2025-09-09T14:29:02.4179182Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[64-dtype6] PASSED 2025-09-09T14:29:02.4180026Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype0] PASSED 2025-09-09T14:29:02.4180834Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype1] PASSED 2025-09-09T14:29:02.4181639Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype2] PASSED 2025-09-09T14:29:02.4182439Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype3] PASSED 2025-09-09T14:29:02.4183246Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype4] PASSED 2025-09-09T14:29:02.4184050Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype5] PASSED 2025-09-09T14:29:02.4184860Z test/dtypes/test_uintx.py::test_uintx_quant_on_cpu_then_move_to_cuda[128-dtype6] PASSED 2025-09-09T14:29:02.4185658Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype0] PASSED 2025-09-09T14:29:02.4186455Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype1] PASSED 2025-09-09T14:29:02.4187212Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype2] PASSED 2025-09-09T14:29:02.4187988Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype3] PASSED 2025-09-09T14:29:02.4188749Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype4] PASSED 2025-09-09T14:29:02.4189489Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype5] PASSED 2025-09-09T14:29:02.4190198Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-32-dtype6] PASSED 2025-09-09T14:29:02.4191114Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype0] PASSED 2025-09-09T14:29:02.4191828Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype1] PASSED 2025-09-09T14:29:02.4192637Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype2] PASSED 2025-09-09T14:29:02.4193350Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype3] PASSED 2025-09-09T14:29:02.4194055Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype4] PASSED 2025-09-09T14:29:02.4194783Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype5] PASSED 2025-09-09T14:29:02.4195583Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-64-dtype6] PASSED 2025-09-09T14:29:02.4196397Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype0] PASSED 2025-09-09T14:29:02.4197135Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype1] PASSED 2025-09-09T14:29:02.4197871Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype2] PASSED 2025-09-09T14:29:02.4198589Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype3] PASSED 2025-09-09T14:29:02.4199320Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype4] PASSED 2025-09-09T14:29:02.4200032Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype5] PASSED 2025-09-09T14:29:02.4200757Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cpu-128-dtype6] PASSED 2025-09-09T14:29:02.4201480Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype0] PASSED 2025-09-09T14:29:02.4202285Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype1] PASSED 2025-09-09T14:29:02.4203006Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype2] PASSED 2025-09-09T14:29:02.4203722Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype3] PASSED 2025-09-09T14:29:02.4204441Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype4] PASSED 2025-09-09T14:29:02.4205166Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype5] PASSED 2025-09-09T14:29:02.4205877Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-32-dtype6] PASSED 2025-09-09T14:29:02.4206601Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype0] PASSED 2025-09-09T14:29:02.4207309Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype1] PASSED 2025-09-09T14:29:02.4208026Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype2] PASSED 2025-09-09T14:29:02.4208740Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype3] PASSED 2025-09-09T14:29:02.4209517Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype4] PASSED 2025-09-09T14:29:02.4210229Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype5] PASSED 2025-09-09T14:29:02.4210941Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-64-dtype6] PASSED 2025-09-09T14:29:02.4211665Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype0] PASSED 2025-09-09T14:29:02.4212385Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype1] PASSED 2025-09-09T14:29:02.4213111Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype2] PASSED 2025-09-09T14:29:02.4213845Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype3] PASSED 2025-09-09T14:29:02.4214562Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype4] PASSED 2025-09-09T14:31:11.6638592Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype5] PASSED 2025-09-09T14:31:11.6642298Z test/dtypes/test_uintx.py::test_uintx_weight_only_model_quant[cuda-128-dtype6] PASSED 2025-09-09T14:31:11.6643041Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype0] PASSED 2025-09-09T14:31:11.6644025Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype1] PASSED 2025-09-09T14:31:11.6644697Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype2] PASSED 2025-09-09T14:31:11.6645393Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype3] PASSED 2025-09-09T14:31:11.6646058Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype4] PASSED 2025-09-09T14:31:11.6646710Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype5] PASSED 2025-09-09T14:31:11.6647364Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-32-dtype6] PASSED 2025-09-09T14:31:11.6648070Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype0] PASSED 2025-09-09T14:31:11.6648717Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype1] PASSED 2025-09-09T14:31:11.6649373Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype2] PASSED 2025-09-09T14:31:11.6650029Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype3] PASSED 2025-09-09T14:31:11.6650687Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype4] PASSED 2025-09-09T14:31:11.6651345Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype5] PASSED 2025-09-09T14:31:11.6651995Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-64-dtype6] PASSED 2025-09-09T14:31:11.6652657Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype0] PASSED 2025-09-09T14:31:11.6653318Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype1] PASSED 2025-09-09T14:31:11.6653992Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype2] PASSED 2025-09-09T14:31:11.6654657Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype3] PASSED 2025-09-09T14:31:11.6655321Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype4] PASSED 2025-09-09T14:31:11.6655981Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype5] PASSED 2025-09-09T14:31:11.6656639Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cpu-128-dtype6] PASSED 2025-09-09T14:31:11.6657305Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype0] PASSED 2025-09-09T14:31:11.6657962Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype1] PASSED 2025-09-09T14:31:11.6658626Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype2] PASSED 2025-09-09T14:31:11.6659288Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype3] PASSED 2025-09-09T14:31:11.6659960Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype4] PASSED 2025-09-09T14:31:11.6660620Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype5] PASSED 2025-09-09T14:31:11.6661281Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-32-dtype6] PASSED 2025-09-09T14:31:11.6661949Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype0] PASSED 2025-09-09T14:31:11.6662614Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype1] PASSED 2025-09-09T14:31:11.6663272Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype2] PASSED 2025-09-09T14:31:11.6664121Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype3] PASSED 2025-09-09T14:31:11.6664782Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype4] PASSED 2025-09-09T14:31:11.6665640Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype5] PASSED 2025-09-09T14:31:11.6666296Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-64-dtype6] PASSED 2025-09-09T14:31:11.6666967Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype0] PASSED 2025-09-09T14:31:11.6667753Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype1] PASSED 2025-09-09T14:31:11.6668421Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype2] PASSED 2025-09-09T14:31:11.6669087Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype3] PASSED 2025-09-09T14:31:11.6669752Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype4] PASSED 2025-09-09T14:31:11.6670423Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype5] PASSED 2025-09-09T14:31:11.6671092Z test/dtypes/test_uintx.py::test_uintx_weight_only_quant[cuda-128-dtype6] PASSED 2025-09-09T14:31:11.6671715Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype0] PASSED 2025-09-09T14:31:11.6672299Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype1] PASSED 2025-09-09T14:31:11.6672869Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype2] PASSED 2025-09-09T14:31:11.6673459Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype3] PASSED 2025-09-09T14:31:11.6674032Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype4] PASSED 2025-09-09T14:31:11.6674608Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype5] PASSED 2025-09-09T14:31:11.6675181Z test/dtypes/test_uintx.py::test_uintx_target_dtype[dtype6] PASSED 2025-09-09T14:31:11.6675777Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype0] PASSED 2025-09-09T14:31:11.6676496Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype1] PASSED 2025-09-09T14:31:11.6677116Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype2] PASSED 2025-09-09T14:31:11.6677740Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype3] PASSED 2025-09-09T14:31:11.6678359Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype4] PASSED 2025-09-09T14:31:11.6678976Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype5] PASSED 2025-09-09T14:31:11.6679594Z test/dtypes/test_uintx.py::test_uintx_target_dtype_compile[dtype6] PASSED 2025-09-09T14:31:11.6680183Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype0] PASSED 2025-09-09T14:31:11.6680751Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype1] PASSED 2025-09-09T14:31:11.6681307Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype2] PASSED 2025-09-09T14:31:11.6681868Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype3] PASSED 2025-09-09T14:31:11.6682428Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype4] PASSED 2025-09-09T14:31:11.6682986Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype5] PASSED 2025-09-09T14:31:11.6683547Z test/dtypes/test_uintx.py::test_uintx_model_size[dtype6] PASSED 2025-09-09T14:31:11.6684349Z test/float8/test_auto_filter.py::test_end_to_end_filtering[tensorwise-module_dims0-valid.layer-filter_fqns0-True] PASSED 2025-09-09T14:31:11.6685390Z test/float8/test_auto_filter.py::test_end_to_end_filtering[tensorwise-module_dims1-skip_layer.linear-filter_fqns1-False] PASSED 2025-09-09T14:31:11.6686661Z test/float8/test_auto_filter.py::test_end_to_end_filtering[tensorwise-module_dims2-valid.layer-filter_fqns2-False] PASSED 2025-09-09T14:31:11.6687836Z test/float8/test_auto_filter.py::test_end_to_end_filtering[rowwise-module_dims3-valid.layer-filter_fqns3-True] PASSED 2025-09-09T14:31:11.6688925Z test/float8/test_auto_filter.py::test_end_to_end_filtering[rowwise-module_dims4-skip_layer.linear-filter_fqns4-False] PASSED 2025-09-09T14:31:11.6690041Z test/float8/test_auto_filter.py::test_end_to_end_filtering[rowwise-module_dims5-valid.layer-filter_fqns5-False] PASSED 2025-09-09T14:31:11.6690863Z test/float8/test_auto_filter.py::test_exact_boundary_dimensions_rowwise PASSED 2025-09-09T14:31:11.6691640Z test/float8/test_auto_filter.py::test_exact_boundary_dimensions_tensorwise PASSED 2025-09-09T14:31:11.6692387Z test/float8/test_auto_filter.py::test_partial_fqn_matching PASSED 2025-09-09T14:31:11.6693141Z test/float8/test_base.py::TestFloat8TrainingTensor::test_preserves_dtype PASSED 2025-09-09T14:31:11.6693965Z test/float8/test_base.py::TestFloat8TrainingTensor::test_differentiable_casts PASSED 2025-09-09T14:31:11.6694765Z test/float8/test_base.py::TestFloat8TrainingTensor::test_split_cat PASSED 2025-09-09T14:31:11.6695515Z test/float8/test_base.py::TestFloat8TrainingTensor::test_index_put PASSED 2025-09-09T14:31:11.6696189Z test/float8/test_base.py::TestFloat8TrainingTensor::test_copy_ PASSED 2025-09-09T14:31:11.6696916Z test/float8/test_base.py::TestFloat8TrainingTensor::test_transpose PASSED 2025-09-09T14:31:11.6697782Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True-0-shape0] PASSED 2025-09-09T14:31:11.6698764Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True-0-shape1] PASSED 2025-09-09T14:31:11.6699758Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True-0-shape2] PASSED 2025-09-09T14:31:11.6700771Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True--1-shape0] PASSED 2025-09-09T14:31:11.6701749Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True--1-shape1] PASSED 2025-09-09T14:31:11.6702724Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[True--1-shape2] PASSED 2025-09-09T14:31:11.6703697Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False-0-shape0] PASSED 2025-09-09T14:31:11.6704684Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False-0-shape1] PASSED 2025-09-09T14:31:11.6705650Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False-0-shape2] PASSED 2025-09-09T14:31:11.6706639Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False--1-shape0] PASSED 2025-09-09T14:31:11.8996503Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False--1-shape1] PASSED 2025-09-09T14:31:11.8997451Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_dynamic_cast[False--1-shape2] PASSED 2025-09-09T14:31:11.8998265Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_reshape PASSED 2025-09-09T14:31:11.8999341Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.AXISWISE-a_shape0] SKIPPED 2025-09-09T14:31:11.9000680Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.AXISWISE-a_shape1] SKIPPED 2025-09-09T14:31:11.9002015Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.AXISWISE-a_shape2] SKIPPED 2025-09-09T14:31:11.9003362Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.TENSORWISE-a_shape0] SKIPPED 2025-09-09T14:31:11.9004720Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.TENSORWISE-a_shape1] SKIPPED 2025-09-09T14:31:11.9006085Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.AXISWISE-ScalingGranularity.TENSORWISE-a_shape2] SKIPPED 2025-09-09T14:31:11.9008186Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.TENSORWISE-ScalingGranularity.AXISWISE-a_shape0] SKIPPED 2025-09-09T14:31:11.9009532Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.TENSORWISE-ScalingGranularity.AXISWISE-a_shape1] SKIPPED 2025-09-09T14:31:11.9010996Z test/float8/test_base.py::TestFloat8TrainingTensor::test_axiswise_gemm[ScalingGranularity.TENSORWISE-ScalingGranularity.AXISWISE-a_shape2] SKIPPED 2025-09-09T14:31:11.9011995Z test/float8/test_base.py::TestFloat8TrainingTensor::test_fp8_dtype PASSED 2025-09-09T14:31:11.9013213Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9014867Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9016513Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9018164Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9019817Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9021451Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9023095Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9024730Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9026360Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9027993Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9029628Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9031254Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[False-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9032887Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9034511Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9036133Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9037865Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9039652Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9041357Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-False-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9043538Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9045250Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9046875Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9048503Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape0-True] PASSED 2025-09-09T14:31:11.9058217Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape1-True] PASSED 2025-09-09T14:31:11.9060043Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_config_params[True-True-linear_dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-x_shape2-True] PASSED 2025-09-09T14:31:11.9061465Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:11.9062699Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:11.9064196Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:11.9065768Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:11.9066995Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:11.9068476Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-True-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:11.9069708Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:11.9070947Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:11.9072176Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:11.9073546Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:11.9074773Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:11.9076101Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype0-False-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8866644Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8868105Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8869583Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8870847Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8872190Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8873431Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-True-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8874663Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8875940Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8877321Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8878569Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8879937Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8881181Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype1-False-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8882428Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8883690Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8884923Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8886140Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8887471Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8888705Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-True-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8890062Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape0-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8891321Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape0-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8892548Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape1-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8893880Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape1-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8895125Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape2-Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:12.8896444Z test/float8/test_base.py::TestFloat8Linear::test_linear_from_recipe[linear_dtype2-False-x_shape2-Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:12.8897790Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.TENSORWISE-linear_dtype0-True] PASSED 2025-09-09T14:31:12.8899038Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.TENSORWISE-linear_dtype1-True] PASSED 2025-09-09T14:31:12.8900161Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.TENSORWISE-linear_dtype2-True] PASSED 2025-09-09T14:31:12.8901369Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE-linear_dtype0-True] PASSED 2025-09-09T14:31:12.8902467Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE-linear_dtype1-True] PASSED 2025-09-09T14:31:12.8903568Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE-linear_dtype2-True] PASSED 2025-09-09T14:31:12.8904721Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE_WITH_GW_HP-linear_dtype0-True] PASSED 2025-09-09T14:31:12.8905893Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE_WITH_GW_HP-linear_dtype1-True] PASSED 2025-09-09T14:31:12.8907302Z test/float8/test_base.py::TestFloat8Linear::test_autocast_outputs[Float8LinearRecipeName.ROWWISE_WITH_GW_HP-linear_dtype2-True] PASSED 2025-09-09T14:31:12.8908318Z test/float8/test_base.py::TestFloat8Linear::test_repr PASSED 2025-09-09T14:31:12.8909022Z test/float8/test_base.py::TestFloat8Linear::test_inference_mode SKIPPED 2025-09-09T14:31:12.8909783Z test/float8/test_base.py::TestFloat8Linear::test_quantize SKIPPED (C...) 2025-09-09T14:31:12.8910662Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[True-base_dtype0] SKIPPED 2025-09-09T14:31:12.8911520Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[True-base_dtype1] SKIPPED 2025-09-09T14:31:12.8912309Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[True-base_dtype2] SKIPPED 2025-09-09T14:31:12.8913104Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[False-base_dtype0] SKIPPED 2025-09-09T14:31:12.8914022Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[False-base_dtype1] SKIPPED 2025-09-09T14:31:12.8914806Z test/float8/test_base.py::TestScaledMM::test_scaled_mm_vs_emulated[False-base_dtype2] SKIPPED 2025-09-09T14:31:12.8915531Z test/float8/test_base.py::TestScaledMM::test_different_configs_error SKIPPED 2025-09-09T14:31:12.8916321Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[True-base_dtype0] SKIPPED 2025-09-09T14:31:12.8917047Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[True-base_dtype1] SKIPPED 2025-09-09T14:31:12.8917876Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[True-base_dtype2] SKIPPED 2025-09-09T14:31:12.8918617Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[False-base_dtype0] SKIPPED 2025-09-09T14:31:12.8919359Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[False-base_dtype1] SKIPPED 2025-09-09T14:31:12.8920084Z test/float8/test_base.py::TestScaledMM::test_pad_inner_dim[False-base_dtype2] SKIPPED 2025-09-09T14:31:12.8920948Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype0] PASSED 2025-09-09T14:31:12.8921681Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype1] PASSED 2025-09-09T14:31:12.8922403Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype2] PASSED 2025-09-09T14:31:12.8923134Z test/float8/test_base.py::TestNumerics::test_small_amax_float16[float8_dtype3] PASSED 2025-09-09T14:31:12.8923982Z test/float8/test_base.py::TestFloat8LinearUtils::test_fp8_tensor_statistics PASSED 2025-09-09T14:31:12.8924835Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_linears_with_filters PASSED 2025-09-09T14:31:12.8925569Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_root_linear PASSED 2025-09-09T14:31:12.8926468Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_root_linear_with_children_raises PASSED 2025-09-09T14:31:12.8927392Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_submodule_linears PASSED 2025-09-09T14:31:12.8928184Z test/float8/test_base.py::TestFloat8LinearUtils::test_swap_submodule_linears_with_skip PASSED 2025-09-09T14:31:12.8929221Z test/float8/test_compile.py::test_eager_only[dtype0-True-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True] PASSED 2025-09-09T14:31:12.8930461Z test/float8/test_compile.py::test_eager_only[dtype1-True-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True] PASSED 2025-09-09T14:31:12.8931772Z test/float8/test_compile.py::test_aot_eager[dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True-True] PASSED 2025-09-09T14:31:12.8933007Z test/float8/test_compile.py::test_aot_eager[dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-True-True] PASSED 2025-09-09T14:31:12.8934405Z test/float8/test_compile.py::test_inductor_from_config_params[dtype0-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-False-True] SKIPPED 2025-09-09T14:31:35.5717318Z test/float8/test_compile.py::test_inductor_from_config_params[dtype1-ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC-False-True] SKIPPED 2025-09-09T14:31:35.5719240Z test/float8/test_compile.py::test_inductor_from_recipe[Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:35.5720182Z test/float8/test_compile.py::test_inductor_from_recipe[Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:35.5721011Z test/float8/test_compile.py::TestGraphBreaks::test_float8_graph_input SKIPPED 2025-09-09T14:31:35.5721698Z test/float8/test_compile.py::TestGraphBreaks::test_float8_graph_output SKIPPED 2025-09-09T14:31:35.5722466Z test/float8/test_compile.py::TestGraphBreaks::test_float8_with_graph_break_in_the_middle SKIPPED 2025-09-09T14:31:35.5723241Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[True-dtype0] SKIPPED 2025-09-09T14:31:35.5723961Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[True-dtype1] SKIPPED 2025-09-09T14:31:35.5724673Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[True-dtype2] SKIPPED 2025-09-09T14:31:35.5725390Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[False-dtype0] SKIPPED 2025-09-09T14:31:35.5726113Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[False-dtype1] SKIPPED 2025-09-09T14:31:35.5726829Z test/float8/test_compile.py::test_dynamic_scale_numeric_parity[False-dtype2] SKIPPED 2025-09-09T14:31:35.5727618Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case0] PASSED 2025-09-09T14:31:35.5728441Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case1] PASSED 2025-09-09T14:31:35.5729254Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case2] PASSED 2025-09-09T14:31:35.5730067Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case3] PASSED 2025-09-09T14:31:35.5730875Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case4] PASSED 2025-09-09T14:31:35.5731694Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case5] PASSED 2025-09-09T14:31:35.5732507Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case6] PASSED 2025-09-09T14:31:35.5734572Z test/float8/test_float8_utils.py::test_round_scale_down_to_power_of_2_valid_inputs[test_case7] PASSED 2025-09-09T14:31:35.5735309Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype0] PASSED 2025-09-09T14:31:35.5736140Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype1] PASSED 2025-09-09T14:31:35.5736809Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype2] PASSED 2025-09-09T14:31:35.5737472Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype3] PASSED 2025-09-09T14:31:35.5738127Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype4] PASSED 2025-09-09T14:31:35.5738791Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype5] PASSED 2025-09-09T14:31:35.5739448Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype6] PASSED 2025-09-09T14:31:35.5740120Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype7] PASSED 2025-09-09T14:31:35.5741330Z test/float8/test_numerics_integration.py::TestFloat8NumericsIntegrationTest::test_encoder_fw_bw_from_config_params[ScalingType.DYNAMIC-ScalingType.DYNAMIC-ScalingType.DYNAMIC] SKIPPED 2025-09-09T14:31:35.5742797Z test/float8/test_numerics_integration.py::TestFloat8NumericsIntegrationTest::test_encoder_fw_bw_from_recipe[Float8LinearRecipeName.ROWWISE] SKIPPED 2025-09-09T14:31:35.5744158Z test/float8/test_numerics_integration.py::TestFloat8NumericsIntegrationTest::test_encoder_fw_bw_from_recipe[Float8LinearRecipeName.ROWWISE_WITH_GW_HP] SKIPPED 2025-09-09T14:31:35.5745104Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_2bit PASSED 2025-09-09T14:31:35.5745671Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_3bit PASSED 2025-09-09T14:31:35.5746230Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_4bit PASSED 2025-09-09T14:31:35.5746792Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_5bit PASSED 2025-09-09T14:31:35.5747351Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_6bit PASSED 2025-09-09T14:31:35.5747902Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_7bit PASSED 2025-09-09T14:31:35.5748467Z test/hqq/test_hqq_affine.py::TestHQQ::test_hqq_plain_8bit PASSED 2025-09-09T14:31:35.5749142Z test/integration/test_integration.py::SmoothquantUnitTest::test_debug_x_absmax PASSED 2025-09-09T14:31:35.5749883Z test/integration/test_integration.py::SmoothquantUnitTest::test_figure_4 PASSED 2025-09-09T14:31:35.5750668Z test/integration/test_integration.py::SmoothquantUnitTest::test_selective_torch_compile PASSED 2025-09-09T14:31:35.5751975Z test/integration/test_integration.py::SmoothquantUnitTest::test_smooth_linear_cpu [W909 14:31:20.837146937 qlinear_dynamic.cpp:251] Warning: Currently, qnnpack incorrectly ignores reduce_range when it is set to true; this may change in a future release. (function operator()) 2025-09-09T14:31:35.5753080Z PASSED 2025-09-09T14:31:35.5753603Z test/integration/test_integration.py::SmoothquantUnitTest::test_smooth_linear_cuda PASSED 2025-09-09T14:31:35.5754436Z test/integration/test_integration.py::SmoothquantUnitTest::test_smooth_linear_edge_cases PASSED 2025-09-09T14:31:35.5755212Z test/integration/test_integration.py::SmoothquantUnitTest::test_swap PASSED 2025-09-09T14:31:35.5755912Z test/integration/test_integration.py::SmoothquantUnitTest::test_tensors PASSED 2025-09-09T14:31:35.5756784Z test/integration/test_integration.py::SmoothquantUnitTest::test_weight_t_and_non_t_numerics_match Autotune Choices Stats: 2025-09-09T14:31:35.5758114Z {"num_choices": 5, "num_triton_choices": 4, "best_kernel": "triton_mm_36", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:31:35.5759205Z AUTOTUNE int_mm(32x32, 32x16) 2025-09-09T14:31:35.5759453Z strides: [32, 1], [1, 32] 2025-09-09T14:31:35.5759692Z dtypes: torch.int8, torch.int8 2025-09-09T14:31:35.5760374Z triton_mm_36 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:31:35.5761330Z triton_mm_38 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=2 2025-09-09T14:31:35.5762268Z triton_mm_39 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:31:35.5763206Z triton_mm_37 0.0256 ms 96.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=2 2025-09-09T14:31:35.5764092Z _int_mm 0.0369 ms 66.7% 2025-09-09T14:31:35.5764552Z SingleProcess AUTOTUNE benchmarking takes 0.0660 seconds and 0.2312 seconds precompiling for 5 choices 2025-09-09T14:31:35.5765100Z PASSED 2025-09-09T14:31:35.5765567Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test__int_mm Autotune Choices Stats: 2025-09-09T14:31:35.5766776Z {"num_choices": 5, "num_triton_choices": 4, "best_kernel": "triton_mm_40", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:31:35.5767775Z AUTOTUNE int_mm(32x32, 32x16) 2025-09-09T14:31:35.5768016Z strides: [32, 1], [16, 1] 2025-09-09T14:31:35.5768264Z dtypes: torch.int8, torch.int8 2025-09-09T14:31:35.5768861Z triton_mm_40 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:31:35.5769819Z triton_mm_41 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=2 2025-09-09T14:31:35.5770765Z triton_mm_42 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=2 2025-09-09T14:31:35.5771699Z triton_mm_43 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:31:35.5772309Z _int_mm 0.0369 ms 69.4% 2025-09-09T14:31:35.5772770Z SingleProcess AUTOTUNE benchmarking takes 0.0660 seconds and 0.2842 seconds precompiling for 5 choices 2025-09-09T14:31:35.5773303Z PASSED 2025-09-09T14:31:35.5773884Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test__int_mm_eager_and_torch_compile_numerics Autotune Choices Stats: 2025-09-09T14:31:35.5775218Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_53", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.07168000191450119, "best_triton_pos": 0} 2025-09-09T14:31:35.5776400Z AUTOTUNE int_mm(17x1536, 1536x1536) 2025-09-09T14:31:35.5776672Z strides: [s15, 1], [s21, 1] 2025-09-09T14:31:35.5776942Z dtypes: torch.int8, torch.int8 2025-09-09T14:31:43.2697065Z triton_mm_53 0.0717 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:31:43.2698081Z triton_mm_56 0.0809 ms 88.6% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:31:43.2699320Z triton_mm_52 0.0932 ms 76.9% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:31:43.2700416Z triton_mm_54 0.0932 ms 76.9% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:31:43.2701369Z triton_mm_48 0.1065 ms 67.3% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:31:43.2702312Z triton_mm_50 0.1075 ms 66.7% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:31:43.2702921Z _int_mm 0.1219 ms 58.8% 2025-09-09T14:31:43.2703530Z triton_mm_58 0.1454 ms 49.3% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:31:43.2704502Z triton_mm_51 0.1608 ms 44.6% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:31:43.2705461Z triton_mm_55 0.1608 ms 44.6% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:31:43.2706314Z SingleProcess AUTOTUNE benchmarking takes 0.2551 seconds and 1.0274 seconds precompiling for 12 choices 2025-09-09T14:31:43.2706832Z Autotune Choices Stats: 2025-09-09T14:31:43.2707764Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_64", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8", "best_time": 0.2170879989862442, "best_triton_pos": 0} 2025-09-09T14:31:43.2708736Z AUTOTUNE int_mm(136x4096, 4096x1536) 2025-09-09T14:31:43.2709009Z strides: [s15, 1], [s21, 1] 2025-09-09T14:31:43.2709264Z dtypes: torch.int8, torch.int8 2025-09-09T14:31:43.2709866Z triton_mm_64 0.2171 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:31:43.2710831Z triton_mm_67 0.2314 ms 93.8% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:31:43.2711779Z triton_mm_59 0.2703 ms 80.3% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:31:43.2712731Z triton_mm_63 0.2939 ms 73.9% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:31:43.2713693Z triton_mm_61 0.3052 ms 71.1% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:31:43.2714309Z _int_mm 0.3144 ms 69.1% 2025-09-09T14:31:43.2714883Z triton_mm_65 0.3512 ms 61.8% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:31:43.2715833Z triton_mm_62 0.4403 ms 49.3% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:31:43.2716885Z triton_mm_60 0.4444 ms 48.8% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:31:43.2717849Z triton_mm_66 0.4966 ms 43.7% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=128, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:31:43.2718799Z SingleProcess AUTOTUNE benchmarking takes 0.4409 seconds and 2.1245 seconds precompiling for 12 choices 2025-09-09T14:31:43.2719501Z PASSED 2025-09-09T14:31:43.2720188Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_dynamic_quant_per_channel_numerics_cpu PASSED 2025-09-09T14:31:43.2721323Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_dynamic_quant_per_channel_numerics_cuda SKIPPED 2025-09-09T14:31:43.2722295Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_per_token_linear_cpu PASSED 2025-09-09T14:31:43.2723193Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_per_token_linear_cuda PASSED 2025-09-09T14:31:43.2724091Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_quantize_per_token_cpu PASSED 2025-09-09T14:31:43.2724999Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_quantize_per_token_cuda PASSED 2025-09-09T14:31:43.2725920Z test/integration/test_integration.py::PythonQuantUtilOpUnitTest::test_quantize_per_token_xpu SKIPPED 2025-09-09T14:31:43.2726892Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_0_cpu SKIPPED 2025-09-09T14:31:43.2727919Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_1_cpu SKIPPED 2025-09-09T14:31:43.2728930Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_2_cpu SKIPPED 2025-09-09T14:31:43.2729953Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_3_cuda SKIPPED 2025-09-09T14:31:43.2730970Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_4_cuda SKIPPED 2025-09-09T14:31:43.2731986Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_rowwise_scaling_subclass_5_cuda SKIPPED 2025-09-09T14:31:43.2733019Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_0_cpu SKIPPED 2025-09-09T14:31:43.2734058Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_1_cpu SKIPPED 2025-09-09T14:31:43.2735101Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_2_cpu SKIPPED 2025-09-09T14:31:43.2736148Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_3_cuda SKIPPED 2025-09-09T14:31:43.2737186Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_4_cuda SKIPPED 2025-09-09T14:31:43.2738231Z test/integration/test_integration.py::TestSubclass::test_aq_float8_dynamic_quant_tensorwise_scaling_subclass_5_cuda SKIPPED 2025-09-09T14:31:43.2739220Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:31:43.2740146Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:31:43.2741076Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:31:43.2742033Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:31:43.2742988Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:31:43.2743921Z test/integration/test_integration.py::TestSubclass::test_aq_float8_weight_only_quant_subclass_5_cuda SKIPPED 2025-09-09T14:31:43.2744815Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_0_cpu SKIPPED 2025-09-09T14:31:43.2745786Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_1_cpu SKIPPED 2025-09-09T14:31:43.2746660Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_2_cpu SKIPPED 2025-09-09T14:31:43.2747554Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_3_cuda Autotune Choices Stats: 2025-09-09T14:31:43.2748804Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_71", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:31:43.2749753Z AUTOTUNE int_mm(32x64, 64x32) 2025-09-09T14:31:43.2750011Z strides: [64, 1], [1, 64] 2025-09-09T14:31:43.2750247Z dtypes: torch.int8, torch.int8 2025-09-09T14:31:43.2750856Z triton_mm_71 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:31:43.2751821Z triton_mm_72 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:32:52.1948023Z triton_mm_73 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:32:52.1950147Z triton_mm_74 0.0256 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:32:52.1951138Z triton_mm_70 0.0266 ms 96.4% ACC_TYPE='tl.int32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1951762Z _int_mm 0.0389 ms 65.8% 2025-09-09T14:32:52.1952250Z SingleProcess AUTOTUNE benchmarking takes 0.0766 seconds and 0.2484 seconds precompiling for 6 choices 2025-09-09T14:32:52.1953000Z PASSED 2025-09-09T14:32:52.1953585Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_4_cuda PASSED 2025-09-09T14:32:52.1954471Z test/integration/test_integration.py::TestSubclass::test_aq_int8_dynamic_quant_subclass_5_cuda PASSED 2025-09-09T14:32:52.1955381Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_0_cpu SKIPPED 2025-09-09T14:32:52.1956460Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_1_cpu SKIPPED 2025-09-09T14:32:52.1957378Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_2_cpu SKIPPED 2025-09-09T14:32:52.1958290Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_3_cuda PASSED 2025-09-09T14:32:52.1959208Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_4_cuda PASSED 2025-09-09T14:32:52.1960121Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_2_subclass_5_cuda PASSED 2025-09-09T14:32:52.1961035Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_0_cpu SKIPPED 2025-09-09T14:32:52.1961953Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_1_cpu SKIPPED 2025-09-09T14:32:52.1962863Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_2_cpu SKIPPED 2025-09-09T14:32:52.1963985Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_3_cuda Autotune Choices Stats: 2025-09-09T14:32:52.1965275Z {"num_choices": 12, "num_triton_choices": 10, "best_kernel": "triton_mm_87", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:32:52.1966564Z AUTOTUNE addmm(32x32, 32x64, 64x32) 2025-09-09T14:32:52.1966831Z strides: [0, 1], [64, 1], [32, 1] 2025-09-09T14:32:52.1967291Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:32:52.1967968Z triton_mm_87 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:32:52.1968942Z triton_mm_92 0.0246 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:32:52.1969903Z triton_mm_85 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:32:52.1970866Z triton_mm_86 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1971828Z triton_mm_88 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:32:52.1972779Z triton_mm_89 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1973735Z triton_mm_94 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:32:52.1974690Z triton_mm_90 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1975645Z triton_mm_91 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:32:52.1976608Z triton_mm_93 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:32:52.1977461Z SingleProcess AUTOTUNE benchmarking takes 0.1548 seconds and 0.3260 seconds precompiling for 12 choices 2025-09-09T14:32:52.1978004Z PASSED 2025-09-09T14:32:52.1978530Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_4_cuda Autotune Choices Stats: 2025-09-09T14:32:52.1979812Z {"num_choices": 12, "num_triton_choices": 10, "best_kernel": "triton_mm_104", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.024607999250292778, "best_triton_pos": 0} 2025-09-09T14:32:52.1980785Z AUTOTUNE addmm(32x32, 32x64, 64x32) 2025-09-09T14:32:52.1981057Z strides: [0, 1], [64, 1], [32, 1] 2025-09-09T14:32:52.1981351Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:32:52.1982039Z triton_mm_104 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:32:52.1983017Z triton_mm_96 0.0256 ms 96.1% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1983973Z triton_mm_97 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:32:52.1984945Z triton_mm_101 0.0266 ms 92.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:32:52.1986055Z triton_mm_102 0.0266 ms 92.5% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:32:52.1987088Z triton_mm_95 0.0266 ms 92.4% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:32:52.1988049Z triton_mm_98 0.0266 ms 92.4% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:32:52.1989007Z triton_mm_99 0.0266 ms 92.4% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1989982Z triton_mm_100 0.0266 ms 92.4% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1990959Z triton_mm_103 0.0266 ms 92.4% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:32:52.1991816Z SingleProcess AUTOTUNE benchmarking takes 0.1570 seconds and 0.3117 seconds precompiling for 12 choices 2025-09-09T14:32:52.1992353Z PASSED 2025-09-09T14:32:52.1992880Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_3_subclass_5_cuda Autotune Choices Stats: 2025-09-09T14:32:52.1994157Z {"num_choices": 12, "num_triton_choices": 10, "best_kernel": "triton_mm_110", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.025536000728607178, "best_triton_pos": 0} 2025-09-09T14:32:52.1995137Z AUTOTUNE addmm(32x32, 32x64, 64x32) 2025-09-09T14:32:52.1995398Z strides: [0, 1], [64, 1], [32, 1] 2025-09-09T14:32:52.1995719Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:32:52.1996514Z triton_mm_110 0.0255 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:32:52.1997500Z triton_mm_114 0.0256 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:32:52.1998472Z triton_mm_108 0.0256 ms 99.6% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:32:52.1999445Z triton_mm_105 0.0266 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:33:05.0908237Z triton_mm_106 0.0266 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:33:05.0909273Z triton_mm_107 0.0266 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:33:05.0910246Z triton_mm_111 0.0266 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:33:05.0911274Z triton_mm_112 0.0266 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:33:05.0912309Z triton_mm_113 0.0266 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:33:05.0913570Z triton_mm_109 0.0287 ms 89.1% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:33:05.0914642Z SingleProcess AUTOTUNE benchmarking takes 0.1571 seconds and 0.3040 seconds precompiling for 12 choices 2025-09-09T14:33:05.0915369Z PASSED 2025-09-09T14:33:05.0915976Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:33:05.0917004Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:33:05.0917905Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:33:05.0918805Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:33:05.0919723Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:33:05.0920621Z test/integration/test_integration.py::TestSubclass::test_aq_int8_weight_only_quant_subclass_5_cuda SKIPPED 2025-09-09T14:33:05.0921518Z test/integration/test_integration.py::TestSubclass::test_autoquantizable_flatten_unflatten PASSED 2025-09-09T14:33:05.0922449Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:33:05.0923418Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:33:05.0924390Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:33:05.0925367Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:33:05.0926345Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:33:05.0927318Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_5_cuda PASSED 2025-09-09T14:33:05.0928326Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_0_cpu SKIPPED 2025-09-09T14:33:05.0929363Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_1_cpu SKIPPED 2025-09-09T14:33:05.0930399Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_2_cpu SKIPPED 2025-09-09T14:33:05.0931431Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_3_cuda SKIPPED 2025-09-09T14:33:05.0932474Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_4_cuda SKIPPED 2025-09-09T14:33:05.0933508Z test/integration/test_integration.py::TestSubclass::test_dequantize_int4_weight_only_quant_subclass_grouped_5_cuda PASSED 2025-09-09T14:33:05.0934546Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_0_cpu PASSED 2025-09-09T14:33:05.0935485Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_1_cpu PASSED 2025-09-09T14:33:05.0936418Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_2_cpu PASSED 2025-09-09T14:33:05.0937365Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_3_cuda PASSED 2025-09-09T14:33:05.0938308Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_4_cuda PASSED 2025-09-09T14:33:05.0939254Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_dynamic_quant_subclass_5_cuda PASSED 2025-09-09T14:33:05.0940318Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_0_cpu PASSED 2025-09-09T14:33:05.0941363Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_1_cpu PASSED 2025-09-09T14:33:05.0942337Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_2_cpu PASSED 2025-09-09T14:33:05.0943309Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_3_cuda PASSED 2025-09-09T14:33:05.0944275Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_4_cuda PASSED 2025-09-09T14:33:05.0945242Z test/integration/test_integration.py::TestSubclass::test_dequantize_int8_weight_only_quant_subclass_5_cuda PASSED 2025-09-09T14:33:05.0946104Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_0_cpu SKIPPED 2025-09-09T14:33:05.0946867Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_1_cpu SKIPPED 2025-09-09T14:33:05.0947635Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_2_cpu SKIPPED 2025-09-09T14:33:05.0948399Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_3_cuda SKIPPED 2025-09-09T14:33:05.0949180Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_4_cuda SKIPPED 2025-09-09T14:33:05.0949940Z test/integration/test_integration.py::TestSubclass::test_gemlite_layout_5_cuda SKIPPED 2025-09-09T14:33:05.0950803Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_0_cpu SKIPPED 2025-09-09T14:33:05.0951753Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_1_cpu SKIPPED 2025-09-09T14:33:05.0952732Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_2_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0953684Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0954242Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0954407Z 2025-09-09T14:33:05.0954577Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0955156Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0955694Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0955861Z 2025-09-09T14:33:05.0956007Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0964231Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0964781Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0964968Z 2025-09-09T14:33:05.0965121Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0965705Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0966242Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0966412Z 2025-09-09T14:33:05.0966572Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0967148Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0967681Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0967843Z 2025-09-09T14:33:05.0967988Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0968569Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0969100Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0969263Z 2025-09-09T14:33:05.0969406Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:05.0970152Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:05.0970673Z return fn(*args, **kwargs) 2025-09-09T14:33:05.0970841Z 2025-09-09T14:33:05.0970990Z PASSED 2025-09-09T14:33:05.0971730Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_3_cuda SKIPPED 2025-09-09T14:33:05.0972693Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_4_cuda SKIPPED 2025-09-09T14:33:05.0973572Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_hqq_quant_subclass_api_5_cuda Autotune Choices Stats: 2025-09-09T14:33:05.0974918Z {"num_choices": 7, "num_triton_choices": 5, "best_kernel": "triton_mm_116", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:33:05.0975900Z AUTOTUNE addmm(16x16, 16x16, 16x16) 2025-09-09T14:33:05.0976166Z strides: [0, 1], [16, 1], [1, 16] 2025-09-09T14:33:05.0976485Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:33:40.4687372Z triton_mm_116 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:33:40.4688386Z triton_mm_117 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:33:40.4689359Z triton_mm_118 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:33:40.4690451Z triton_mm_119 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:33:40.4691422Z triton_mm_115 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:33:40.4692043Z addmm 0.0502 ms 49.0% 2025-09-09T14:33:40.4692292Z bias_addmm 0.0707 ms 34.8% 2025-09-09T14:33:40.4692770Z SingleProcess AUTOTUNE benchmarking takes 0.0932 seconds and 0.2681 seconds precompiling for 7 choices 2025-09-09T14:33:40.4693505Z PASSED 2025-09-09T14:33:40.4694098Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:33:40.4694993Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:33:40.4695873Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:33:40.4696820Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_3_cuda SKIPPED 2025-09-09T14:33:40.4697706Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_4_cuda SKIPPED 2025-09-09T14:33:40.4698588Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_5_cuda PASSED 2025-09-09T14:33:40.4699490Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_0_cpu SKIPPED 2025-09-09T14:33:40.4700400Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_1_cpu SKIPPED 2025-09-09T14:33:40.4701312Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_2_cpu PASSED 2025-09-09T14:33:40.4702231Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_3_cuda SKIPPED 2025-09-09T14:33:40.4703143Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_4_cuda SKIPPED 2025-09-09T14:33:40.4704385Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_5_cuda PASSED 2025-09-09T14:33:40.4705500Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_0_cpu SKIPPED 2025-09-09T14:33:40.4706499Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_1_cpu SKIPPED 2025-09-09T14:33:40.4707501Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_2_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:40.4708458Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:40.4708996Z return fn(*args, **kwargs) 2025-09-09T14:33:40.4709166Z 2025-09-09T14:33:40.4709323Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:40.4709911Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:40.4710440Z return fn(*args, **kwargs) 2025-09-09T14:33:40.4710602Z 2025-09-09T14:33:40.4710746Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:40.4711328Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:40.4711857Z return fn(*args, **kwargs) 2025-09-09T14:33:40.4712017Z 2025-09-09T14:33:40.4712160Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:40.4712732Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:40.4713256Z return fn(*args, **kwargs) 2025-09-09T14:33:40.4713422Z 2025-09-09T14:33:40.4713565Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:40.4714134Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:40.4714667Z return fn(*args, **kwargs) 2025-09-09T14:33:40.4714825Z 2025-09-09T14:33:40.4714974Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:33:40.4715548Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:33:40.4716074Z return fn(*args, **kwargs) 2025-09-09T14:33:40.4716237Z 2025-09-09T14:33:40.4716449Z PASSED 2025-09-09T14:33:40.4717075Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_3_cuda SKIPPED 2025-09-09T14:33:40.4718056Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_4_cuda SKIPPED 2025-09-09T14:33:40.4718966Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_api_grouped_5_cuda Autotune Choices Stats: 2025-09-09T14:33:40.4720296Z {"num_choices": 13, "num_triton_choices": 11, "best_kernel": "triton_mm_129", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:33:40.4721295Z AUTOTUNE addmm(256x16, 256x16, 16x16) 2025-09-09T14:33:40.4721566Z strides: [0, 1], [16, 1], [1, 16] 2025-09-09T14:33:40.4721885Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:33:40.4722581Z triton_mm_129 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:33:40.4723579Z triton_mm_130 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:33:40.4724566Z triton_mm_121 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:33:40.4725624Z triton_mm_122 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:33:40.4726721Z triton_mm_126 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:33:40.4727697Z triton_mm_127 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:33:40.4728671Z triton_mm_128 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:33:40.4729646Z triton_mm_124 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:33:40.4730607Z triton_mm_125 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:33:40.4731572Z triton_mm_120 0.0276 ms 88.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:33:40.4732425Z SingleProcess AUTOTUNE benchmarking takes 0.1688 seconds and 0.2945 seconds precompiling for 13 choices 2025-09-09T14:33:40.4732927Z Autotune Choices Stats: 2025-09-09T14:33:40.4733858Z {"num_choices": 13, "num_triton_choices": 11, "best_kernel": "triton_mm_169", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:33:40.4734830Z AUTOTUNE addmm(256x8, 256x8, 8x8) 2025-09-09T14:33:40.4735087Z strides: [0, 1], [8, 1], [1, 8] 2025-09-09T14:33:40.4735394Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:33:40.4736086Z triton_mm_169 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:33:40.4737074Z triton_mm_170 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:33:40.4738059Z triton_mm_173 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:33:40.4739054Z triton_mm_174 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:33:40.4740045Z triton_mm_171 0.0246 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:07.9137364Z triton_mm_172 0.0246 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:34:07.9138536Z triton_mm_166 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:34:07.9139616Z triton_mm_167 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:07.9140696Z triton_mm_168 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:07.9142172Z triton_mm_164 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:34:07.9143331Z SingleProcess AUTOTUNE benchmarking takes 0.1701 seconds and 0.2852 seconds precompiling for 13 choices 2025-09-09T14:34:07.9144153Z PASSED 2025-09-09T14:34:07.9144871Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_0_cpu SKIPPED 2025-09-09T14:34:07.9145928Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_1_cpu SKIPPED 2025-09-09T14:34:07.9146965Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_2_cpu SKIPPED 2025-09-09T14:34:07.9148058Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_3_cuda SKIPPED 2025-09-09T14:34:07.9149144Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_4_cuda SKIPPED 2025-09-09T14:34:07.9150198Z test/integration/test_integration.py::TestSubclass::test_int4_weight_only_quant_subclass_grouped_5_cuda SKIPPED 2025-09-09T14:34:07.9151194Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_0_cpu SKIPPED 2025-09-09T14:34:07.9152142Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_1_cpu SKIPPED 2025-09-09T14:34:07.9153089Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_2_cpu SKIPPED 2025-09-09T14:34:07.9154033Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_3_cuda PASSED 2025-09-09T14:34:07.9154994Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_4_cuda PASSED 2025-09-09T14:34:07.9155970Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_5_cuda PASSED 2025-09-09T14:34:07.9157028Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_00_cpu SKIPPED 2025-09-09T14:34:07.9158028Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_01_cpu SKIPPED 2025-09-09T14:34:07.9159019Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_02_cpu SKIPPED 2025-09-09T14:34:07.9160000Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_03_cpu SKIPPED 2025-09-09T14:34:07.9160991Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_04_cpu SKIPPED 2025-09-09T14:34:07.9161978Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_05_cpu SKIPPED 2025-09-09T14:34:07.9163025Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_06_cuda SKIPPED 2025-09-09T14:34:07.9164263Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_07_cuda SKIPPED 2025-09-09T14:34:07.9165291Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_08_cuda SKIPPED 2025-09-09T14:34:07.9166288Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_09_cuda SKIPPED 2025-09-09T14:34:07.9167278Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_10_cuda SKIPPED 2025-09-09T14:34:07.9168274Z test/integration/test_integration.py::TestSubclass::test_int8_dynamic_quant_subclass_api_11_cuda SKIPPED 2025-09-09T14:34:07.9169258Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_0_cpu SKIPPED 2025-09-09T14:34:07.9170357Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_1_cpu SKIPPED 2025-09-09T14:34:07.9171260Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_2_cpu SKIPPED 2025-09-09T14:34:07.9172326Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_3_cuda Autotune Choices Stats: 2025-09-09T14:34:07.9173639Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_229", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:34:07.9174687Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:34:07.9174923Z strides: [64, 1], [32, 1] 2025-09-09T14:34:07.9175181Z dtypes: torch.float32, torch.float32 2025-09-09T14:34:07.9175911Z triton_mm_229 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:07.9176931Z triton_mm_232 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:34:07.9177936Z triton_mm_226 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:07.9178942Z triton_mm_228 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:07.9179926Z triton_mm_230 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:07.9180915Z triton_mm_231 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:34:07.9181903Z triton_mm_224 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:07.9182990Z triton_mm_225 0.0276 ms 88.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:07.9183976Z triton_mm_223 0.0287 ms 85.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:34:07.9184956Z triton_mm_227 0.0287 ms 85.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:07.9185829Z SingleProcess AUTOTUNE benchmarking takes 0.1435 seconds and 0.4190 seconds precompiling for 11 choices 2025-09-09T14:34:07.9186392Z PASSED 2025-09-09T14:34:07.9186906Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_4_cuda Autotune Choices Stats: 2025-09-09T14:34:07.9188200Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_240", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:34:07.9189180Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:34:07.9189422Z strides: [64, 1], [32, 1] 2025-09-09T14:34:07.9189688Z dtypes: torch.float16, torch.float16 2025-09-09T14:34:07.9190333Z triton_mm_240 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:07.9191434Z triton_mm_242 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:34:07.9192502Z triton_mm_233 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:34:07.9193486Z triton_mm_234 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:07.9194471Z triton_mm_235 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:07.9195449Z triton_mm_236 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:32.6855749Z triton_mm_237 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:32.6856788Z triton_mm_238 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:32.6857771Z triton_mm_239 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:32.6858745Z triton_mm_241 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:34:32.6859604Z SingleProcess AUTOTUNE benchmarking takes 0.1394 seconds and 0.3041 seconds precompiling for 11 choices 2025-09-09T14:34:32.6860381Z PASSED 2025-09-09T14:34:32.6860945Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_5_cuda Autotune Choices Stats: 2025-09-09T14:34:32.6862223Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_250", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:34:32.6863203Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:34:32.6863442Z strides: [64, 1], [32, 1] 2025-09-09T14:34:32.6863856Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:34:32.6864511Z triton_mm_250 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:32.6865488Z triton_mm_243 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:34:32.6866471Z triton_mm_244 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:32.6867447Z triton_mm_245 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:32.6868411Z triton_mm_246 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:32.6869379Z triton_mm_248 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:32.6870354Z triton_mm_249 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:34:32.6871678Z triton_mm_251 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:34:32.6872818Z triton_mm_252 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:34:32.6873789Z triton_mm_247 0.0257 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:32.6874641Z SingleProcess AUTOTUNE benchmarking takes 0.1392 seconds and 0.2915 seconds precompiling for 11 choices 2025-09-09T14:34:32.6875185Z PASSED 2025-09-09T14:34:32.6875798Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_0_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6876807Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6877342Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6877506Z 2025-09-09T14:34:32.6877657Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6878247Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6878771Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6878937Z 2025-09-09T14:34:32.6879082Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6879664Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6880184Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6880344Z 2025-09-09T14:34:32.6880491Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6881058Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6881589Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6881748Z 2025-09-09T14:34:32.6881890Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6882470Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6883001Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6883160Z 2025-09-09T14:34:32.6883302Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6883876Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6884396Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6884562Z 2025-09-09T14:34:32.6884680Z PASSED 2025-09-09T14:34:32.6885301Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_1_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6886223Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6886753Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6886911Z 2025-09-09T14:34:32.6887053Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6887633Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6888158Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6888324Z 2025-09-09T14:34:32.6888466Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6889042Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6889561Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6889726Z 2025-09-09T14:34:32.6889868Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6890434Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6891054Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6891212Z 2025-09-09T14:34:32.6891360Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6892004Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6892539Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6892701Z 2025-09-09T14:34:32.6892842Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6893422Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6893939Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6894107Z 2025-09-09T14:34:32.6894227Z PASSED 2025-09-09T14:34:32.6894844Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_2_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6895771Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6896299Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6896457Z 2025-09-09T14:34:32.6896598Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6897180Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6897705Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6897867Z 2025-09-09T14:34:32.6898010Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6898581Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6899097Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6899264Z 2025-09-09T14:34:32.6899406Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:34:32.6900019Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torch/_dynamo/external_utils.py", line 68, in inner 2025-09-09T14:34:32.6900554Z return fn(*args, **kwargs) 2025-09-09T14:34:32.6900712Z 2025-09-09T14:34:32.6900834Z PASSED 2025-09-09T14:34:32.6901349Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_3_cuda Autotune Choices Stats: 2025-09-09T14:34:32.6902644Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_263", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:34:32.6903611Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:34:32.6903838Z strides: [32, 1], [32, 1] 2025-09-09T14:34:32.6904088Z dtypes: torch.float32, torch.float32 2025-09-09T14:34:32.6904723Z triton_mm_263 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:34:32.6905717Z triton_mm_265 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:34:32.6906700Z triton_mm_266 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:34:32.6907675Z triton_mm_267 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:19.5985171Z triton_mm_268 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:19.5986205Z triton_mm_264 0.0266 ms 96.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:19.5987182Z mm 0.0379 ms 67.6% 2025-09-09T14:35:19.5987645Z SingleProcess AUTOTUNE benchmarking takes 0.0888 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:35:19.5988362Z PASSED 2025-09-09T14:35:19.5989098Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_4_cuda Autotune Choices Stats: 2025-09-09T14:35:19.5990399Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_279", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:35:19.5991394Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:35:19.5991627Z strides: [32, 1], [32, 1] 2025-09-09T14:35:19.5991879Z dtypes: torch.float16, torch.float16 2025-09-09T14:35:19.5992524Z triton_mm_279 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:19.5993536Z triton_mm_280 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:19.5994539Z triton_mm_282 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:19.5995527Z triton_mm_283 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:19.5996595Z triton_mm_284 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:19.5997584Z triton_mm_281 0.0257 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:19.5998212Z mm 0.0399 ms 64.1% 2025-09-09T14:35:19.5998672Z SingleProcess AUTOTUNE benchmarking takes 0.0881 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:35:19.5999213Z PASSED 2025-09-09T14:35:19.5999763Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_subclass_api_5_cuda Autotune Choices Stats: 2025-09-09T14:35:19.6001094Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_297", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:35:19.6002061Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:35:19.6002305Z strides: [32, 1], [32, 1] 2025-09-09T14:35:19.6002559Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:35:19.6003219Z triton_mm_297 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:19.6004221Z triton_mm_295 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:19.6005198Z triton_mm_296 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:19.6006179Z triton_mm_298 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:19.6007160Z triton_mm_299 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:19.6008229Z triton_mm_300 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:19.6008852Z mm 0.0389 ms 63.2% 2025-09-09T14:35:19.6009378Z SingleProcess AUTOTUNE benchmarking takes 0.0872 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:35:19.6009919Z PASSED 2025-09-09T14:35:19.6010436Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_0_cpu Autotune Choices Stats: 2025-09-09T14:35:19.6011245Z {"num_choices": 2, "num_triton_choices": 0, "best_kernel": "cpp_CppMicroGemmFP32Vec_0", "best_time": 0.007110000296961516} 2025-09-09T14:35:19.6011788Z AUTOTUNE packed_linear(32x64, 1459233x1, 32x64) 2025-09-09T14:35:19.6012095Z strides: [64, 1], [1, 0], [64, 1] 2025-09-09T14:35:19.6012400Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:35:19.6012750Z cpp_CppMicroGemmFP32Vec_0 0.0071 ms 100.0% 2025-09-09T14:35:19.6013051Z _mkl_linear 0.0266 ms 26.7% 2025-09-09T14:35:19.6013527Z SingleProcess AUTOTUNE benchmarking takes 0.2493 seconds and 2.1467 seconds precompiling for 2 choices 2025-09-09T14:35:19.6014027Z Autotune Choices Stats: 2025-09-09T14:35:19.6014510Z {"num_choices": 2, "num_triton_choices": 0, "best_kernel": "cpp_CppMicroGemmFP32Vec_1", "best_time": 0.005909999799769139} 2025-09-09T14:35:19.6015048Z AUTOTUNE packed_linear(32x32, 1459233x1, 32x32) 2025-09-09T14:35:19.6015351Z strides: [32, 1], [1, 0], [32, 1] 2025-09-09T14:35:19.6015649Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:35:19.6016012Z cpp_CppMicroGemmFP32Vec_1 0.0059 ms 100.0% 2025-09-09T14:35:19.6016316Z _mkl_linear 0.0263 ms 22.5% 2025-09-09T14:35:19.6016784Z SingleProcess AUTOTUNE benchmarking takes 0.2491 seconds and 2.1503 seconds precompiling for 2 choices 2025-09-09T14:35:19.6017313Z PASSED 2025-09-09T14:35:19.6017830Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_1_cpu Autotune Choices Stats: 2025-09-09T14:35:19.6018637Z {"num_choices": 2, "num_triton_choices": 0, "best_kernel": "cpp_CppMicroGemmFP32Vec_2", "best_time": 0.006329999905574368} 2025-09-09T14:35:19.6019148Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:35:19.6019385Z strides: [64, 1], [1, 64] 2025-09-09T14:35:19.6019640Z dtypes: torch.float16, torch.float16 2025-09-09T14:35:19.6019940Z cpp_CppMicroGemmFP32Vec_2 0.0063 ms 100.0% 2025-09-09T14:35:19.6020227Z mm 0.0312 ms 20.3% 2025-09-09T14:35:19.6020672Z SingleProcess AUTOTUNE benchmarking takes 0.2541 seconds and 2.2967 seconds precompiling for 2 choices 2025-09-09T14:35:19.6021173Z Autotune Choices Stats: 2025-09-09T14:35:19.6021640Z {"num_choices": 2, "num_triton_choices": 0, "best_kernel": "cpp_CppMicroGemmFP32Vec_3", "best_time": 0.006199999916134402} 2025-09-09T14:35:19.6022143Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:35:19.6022384Z strides: [32, 1], [1, 32] 2025-09-09T14:35:19.6022632Z dtypes: torch.float16, torch.float16 2025-09-09T14:35:19.6022937Z cpp_CppMicroGemmFP32Vec_3 0.0062 ms 100.0% 2025-09-09T14:35:19.6023225Z mm 0.0311 ms 19.9% 2025-09-09T14:35:19.6023675Z SingleProcess AUTOTUNE benchmarking takes 0.2543 seconds and 2.2990 seconds precompiling for 2 choices 2025-09-09T14:35:19.6024203Z PASSED 2025-09-09T14:35:19.6024722Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_2_cpu Autotune Choices Stats: 2025-09-09T14:35:19.6025532Z {"num_choices": 2, "num_triton_choices": 0, "best_kernel": "cpp_CppMicroGemmFP32Vec_4", "best_time": 0.006471000006058603} 2025-09-09T14:35:19.6026068Z AUTOTUNE _weight_int8pack_mm(32x64, 32x64, 32) 2025-09-09T14:35:19.6026375Z strides: [64, 1], [64, 1], [1] 2025-09-09T14:35:19.6026662Z dtypes: torch.bfloat16, torch.int8, torch.bfloat16 2025-09-09T14:35:19.6027011Z cpp_CppMicroGemmFP32Vec_4 0.0065 ms 100.0% 2025-09-09T14:35:19.6027315Z _weight_int8pack_mm 0.0176 ms 36.8% 2025-09-09T14:35:19.6027910Z SingleProcess AUTOTUNE benchmarking takes 0.2498 seconds and 2.2735 seconds precompiling for 2 choices 2025-09-09T14:35:19.6028416Z Autotune Choices Stats: 2025-09-09T14:35:19.6028882Z {"num_choices": 2, "num_triton_choices": 0, "best_kernel": "cpp_CppMicroGemmFP32Vec_5", "best_time": 0.006309999662335031} 2025-09-09T14:35:19.6029500Z AUTOTUNE _weight_int8pack_mm(32x32, 32x32, 32) 2025-09-09T14:35:19.6029798Z strides: [32, 1], [32, 1], [1] 2025-09-09T14:35:19.6030125Z dtypes: torch.bfloat16, torch.int8, torch.bfloat16 2025-09-09T14:35:19.6030488Z cpp_CppMicroGemmFP32Vec_5 0.0063 ms 100.0% 2025-09-09T14:35:19.6030796Z _weight_int8pack_mm 0.0173 ms 36.4% 2025-09-09T14:35:19.6031289Z SingleProcess AUTOTUNE benchmarking takes 0.2498 seconds and 2.2855 seconds precompiling for 2 choices 2025-09-09T14:35:19.6031822Z PASSED 2025-09-09T14:35:19.6032342Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_3_cuda Autotune Choices Stats: 2025-09-09T14:35:19.6033632Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_307", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:35:19.6034611Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:35:19.6034853Z strides: [64, 1], [1, 64] 2025-09-09T14:35:19.6035105Z dtypes: torch.float32, torch.float32 2025-09-09T14:35:19.6035754Z triton_mm_307 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:19.6036793Z triton_mm_304 0.0236 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:19.6037794Z triton_mm_301 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:36.4129478Z triton_mm_302 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4130572Z triton_mm_303 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:36.4131608Z triton_mm_305 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4132591Z triton_mm_306 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4133571Z triton_mm_308 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:36.4134553Z triton_mm_309 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:36.4135537Z triton_mm_310 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:36.4136406Z SingleProcess AUTOTUNE benchmarking takes 0.1387 seconds and 0.3616 seconds precompiling for 11 choices 2025-09-09T14:35:36.4136926Z Autotune Choices Stats: 2025-09-09T14:35:36.4137871Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_311", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:35:36.4139192Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:35:36.4139432Z strides: [32, 1], [1, 32] 2025-09-09T14:35:36.4139685Z dtypes: torch.float32, torch.float32 2025-09-09T14:35:36.4140499Z triton_mm_311 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:36.4141493Z triton_mm_312 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4142485Z triton_mm_313 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:36.4143476Z triton_mm_314 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4144462Z triton_mm_315 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:36.4145453Z triton_mm_316 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:36.4146077Z mm 0.0379 ms 67.6% 2025-09-09T14:35:36.4146525Z SingleProcess AUTOTUNE benchmarking takes 0.0885 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:35:36.4147243Z PASSED 2025-09-09T14:35:36.4147788Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_4_cuda Autotune Choices Stats: 2025-09-09T14:35:36.4149084Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_317", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:35:36.4150064Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:35:36.4150294Z strides: [64, 1], [1, 64] 2025-09-09T14:35:36.4150548Z dtypes: torch.float16, torch.float16 2025-09-09T14:35:36.4151190Z triton_mm_317 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:36.4152187Z triton_mm_318 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4153169Z triton_mm_319 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:36.4154138Z triton_mm_321 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4155123Z triton_mm_322 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4156102Z triton_mm_323 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:36.4157211Z triton_mm_325 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:36.4158191Z triton_mm_326 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:36.4159265Z triton_mm_324 0.0246 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:36.4160320Z triton_mm_320 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:36.4161195Z SingleProcess AUTOTUNE benchmarking takes 0.1406 seconds and 0.2106 seconds precompiling for 11 choices 2025-09-09T14:35:36.4161695Z Autotune Choices Stats: 2025-09-09T14:35:36.4162634Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_327", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:35:36.4163600Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:35:36.4164104Z strides: [32, 1], [1, 32] 2025-09-09T14:35:36.4164367Z dtypes: torch.float16, torch.float16 2025-09-09T14:35:36.4165006Z triton_mm_327 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:36.4166003Z triton_mm_329 0.0255 ms 96.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:36.4166982Z triton_mm_331 0.0255 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:36.4167956Z triton_mm_328 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4168931Z triton_mm_330 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4169914Z triton_mm_332 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:36.4170532Z mm 0.0389 ms 63.2% 2025-09-09T14:35:36.4170988Z SingleProcess AUTOTUNE benchmarking takes 0.0940 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:35:36.4171529Z PASSED 2025-09-09T14:35:36.4172052Z test/integration/test_integration.py::TestSubclass::test_int8_weight_only_quant_with_freeze_5_cuda Autotune Choices Stats: 2025-09-09T14:35:36.4173338Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_333", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:35:36.4174306Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:35:36.4174540Z strides: [64, 1], [32, 1] 2025-09-09T14:35:36.4174789Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:35:36.4175455Z triton_mm_333 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:36.4176453Z triton_mm_334 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:36.4177434Z triton_mm_336 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:55.7037836Z triton_mm_335 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:55.7039181Z triton_mm_337 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:55.7040337Z triton_mm_338 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:55.7041321Z triton_mm_341 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:55.7042294Z triton_mm_342 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:55.7043324Z triton_mm_339 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:55.7044310Z triton_mm_340 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:55.7045181Z SingleProcess AUTOTUNE benchmarking takes 0.1416 seconds and 0.2230 seconds precompiling for 11 choices 2025-09-09T14:35:55.7045697Z Autotune Choices Stats: 2025-09-09T14:35:55.7046645Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_346", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:35:55.7047685Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:35:55.7047923Z strides: [32, 1], [32, 1] 2025-09-09T14:35:55.7048176Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:35:55.7048832Z triton_mm_346 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:55.7049832Z triton_mm_345 0.0246 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:55.7050810Z triton_mm_343 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:55.7051787Z triton_mm_344 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:55.7052768Z triton_mm_347 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:55.7053742Z triton_mm_348 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:55.7054369Z mm 0.0399 ms 61.5% 2025-09-09T14:35:55.7054818Z SingleProcess AUTOTUNE benchmarking takes 0.0879 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:35:55.7055536Z PASSED 2025-09-09T14:35:55.7056057Z test/integration/test_integration.py::TestDynamicQuant::test_dynamic_quant PASSED 2025-09-09T14:35:55.7056943Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_groupwise_embedding_quant PASSED 2025-09-09T14:35:55.7057917Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_groupwise_quant PASSED 2025-09-09T14:35:55.7058796Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant PASSED 2025-09-09T14:35:55.7059738Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_0_cpu SKIPPED 2025-09-09T14:35:55.7060853Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_1_cpu SKIPPED 2025-09-09T14:35:55.7061854Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_2_cpu SKIPPED 2025-09-09T14:35:55.7062872Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_3_cuda Autotune Choices Stats: 2025-09-09T14:35:55.7064567Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_351", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:35:55.7065539Z AUTOTUNE mm(2x4, 4x5) 2025-09-09T14:35:55.7065768Z strides: [4, 1], [5, 1] 2025-09-09T14:35:55.7066009Z dtypes: torch.float32, torch.float32 2025-09-09T14:35:55.7066666Z triton_mm_351 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:35:55.7067672Z triton_mm_352 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:35:55.7068661Z triton_mm_349 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:35:55.7069645Z triton_mm_350 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:35:55.7070626Z triton_mm_353 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:35:55.7071250Z mm 0.0359 ms 68.5% 2025-09-09T14:35:55.7071704Z SingleProcess AUTOTUNE benchmarking takes 0.0769 seconds and 0.2938 seconds precompiling for 6 choices 2025-09-09T14:35:55.7072199Z Autotune Choices Stats: 2025-09-09T14:35:55.7073148Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_356", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:35:55.7074116Z AUTOTUNE mm(125x4, 4x5) 2025-09-09T14:35:55.7074334Z strides: [4, 1], [5, 1] 2025-09-09T14:35:55.7074571Z dtypes: torch.float32, torch.float32 2025-09-09T14:35:55.7075213Z triton_mm_356 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:35:55.7076209Z triton_mm_354 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:35:55.7077266Z triton_mm_355 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:35:55.7078244Z triton_mm_357 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:35:55.7079228Z triton_mm_358 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:35:55.7080206Z triton_mm_359 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:55.7081346Z triton_mm_360 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:35:55.7082451Z triton_mm_361 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:35:55.7083496Z triton_mm_362 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:35:55.7084491Z triton_mm_363 0.0256 ms 88.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:35:55.7085363Z SingleProcess AUTOTUNE benchmarking takes 0.1504 seconds and 0.3312 seconds precompiling for 12 choices 2025-09-09T14:35:55.7085860Z Autotune Choices Stats: 2025-09-09T14:35:55.7086804Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_367", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:35:55.7087774Z AUTOTUNE mm(4x4, 4x5) 2025-09-09T14:35:55.7087991Z strides: [4, 1], [5, 1] 2025-09-09T14:35:55.7088233Z dtypes: torch.float32, torch.float32 2025-09-09T14:36:21.5483531Z triton_mm_367 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:36:21.5485947Z triton_mm_366 0.0256 ms 96.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:36:21.5487497Z triton_mm_365 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:36:21.5488541Z triton_mm_368 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:36:21.5489526Z triton_mm_369 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:36:21.5490145Z mm 0.0369 ms 66.7% 2025-09-09T14:36:21.5490595Z SingleProcess AUTOTUNE benchmarking takes 0.0764 seconds and 0.3033 seconds precompiling for 6 choices 2025-09-09T14:36:21.5491316Z PASSED 2025-09-09T14:36:21.5491906Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_4_cuda Autotune Choices Stats: 2025-09-09T14:36:21.5493239Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_373", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:36:21.5494227Z AUTOTUNE mm(2x4, 4x5) 2025-09-09T14:36:21.5494447Z strides: [4, 1], [5, 1] 2025-09-09T14:36:21.5494687Z dtypes: torch.float16, torch.float16 2025-09-09T14:36:21.5495336Z triton_mm_373 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:36:21.5496669Z triton_mm_374 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:36:21.5497665Z triton_mm_370 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:36:21.5499022Z triton_mm_371 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:36:21.5500157Z triton_mm_372 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:36:21.5500781Z mm 0.0440 ms 53.5% 2025-09-09T14:36:21.5501227Z SingleProcess AUTOTUNE benchmarking takes 0.0769 seconds and 0.2895 seconds precompiling for 6 choices 2025-09-09T14:36:21.5501735Z Autotune Choices Stats: 2025-09-09T14:36:21.5502687Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_385", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:36:21.5503657Z AUTOTUNE mm(125x4, 4x5) 2025-09-09T14:36:21.5503878Z strides: [4, 1], [5, 1] 2025-09-09T14:36:21.5504111Z dtypes: torch.float16, torch.float16 2025-09-09T14:36:21.5504767Z triton_mm_385 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:36:21.5505774Z triton_mm_384 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:36:21.5506754Z triton_mm_375 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:36:21.5507735Z triton_mm_376 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:36:21.5508716Z triton_mm_377 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:36:21.5509692Z triton_mm_378 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:36:21.5510671Z triton_mm_379 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:36:21.5511653Z triton_mm_380 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:36:21.5512621Z triton_mm_381 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:36:21.5513607Z triton_mm_382 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:36:21.5514480Z SingleProcess AUTOTUNE benchmarking takes 0.1547 seconds and 0.2901 seconds precompiling for 12 choices 2025-09-09T14:36:21.5514991Z Autotune Choices Stats: 2025-09-09T14:36:21.5515917Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_387", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1", "best_time": 0.024639999493956566, "best_triton_pos": 0} 2025-09-09T14:36:21.5516965Z AUTOTUNE mm(4x4, 4x5) 2025-09-09T14:36:21.5517186Z strides: [4, 1], [5, 1] 2025-09-09T14:36:21.5517427Z dtypes: torch.float16, torch.float16 2025-09-09T14:36:21.5518070Z triton_mm_387 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:36:21.5519154Z triton_mm_386 0.0256 ms 96.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:36:21.5520214Z triton_mm_388 0.0256 ms 96.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:36:21.5521196Z triton_mm_389 0.0256 ms 96.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:36:21.5522169Z triton_mm_390 0.0256 ms 96.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:36:21.5522791Z mm 0.0451 ms 54.7% 2025-09-09T14:36:21.5523247Z SingleProcess AUTOTUNE benchmarking takes 0.0786 seconds and 0.2879 seconds precompiling for 6 choices 2025-09-09T14:36:21.5523802Z PASSED 2025-09-09T14:36:21.5524379Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_force_mixed_mm_5_cuda Autotune Choices Stats: 2025-09-09T14:36:21.5525703Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_392", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:36:21.5526668Z AUTOTUNE mm(2x4, 4x5) 2025-09-09T14:36:21.5526889Z strides: [4, 1], [5, 1] 2025-09-09T14:36:21.5527130Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:36:21.5527778Z triton_mm_392 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:36:21.5528768Z triton_mm_393 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:36:21.5529760Z triton_mm_391 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:36:21.5530743Z triton_mm_394 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:36:21.5531713Z triton_mm_395 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:36:21.5532331Z mm 0.0389 ms 63.2% 2025-09-09T14:36:21.5532777Z SingleProcess AUTOTUNE benchmarking takes 0.0786 seconds and 0.2891 seconds precompiling for 6 choices 2025-09-09T14:36:21.5533280Z Autotune Choices Stats: 2025-09-09T14:37:14.2069507Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_398", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:37:14.2071608Z AUTOTUNE mm(125x4, 4x5) 2025-09-09T14:37:14.2072060Z strides: [4, 1], [5, 1] 2025-09-09T14:37:14.2072555Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:37:14.2073404Z triton_mm_398 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:37:14.2074425Z triton_mm_399 0.0246 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:14.2075819Z triton_mm_396 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:14.2077052Z triton_mm_400 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:14.2078060Z triton_mm_401 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:14.2079062Z triton_mm_402 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:14.2080068Z triton_mm_403 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:14.2081092Z triton_mm_404 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:37:14.2082110Z triton_mm_405 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:37:14.2083127Z triton_mm_406 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=128, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:37:14.2084018Z SingleProcess AUTOTUNE benchmarking takes 0.1550 seconds and 0.2951 seconds precompiling for 12 choices 2025-09-09T14:37:14.2084535Z Autotune Choices Stats: 2025-09-09T14:37:14.2085493Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_408", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:37:14.2086484Z AUTOTUNE mm(4x4, 4x5) 2025-09-09T14:37:14.2086702Z strides: [4, 1], [5, 1] 2025-09-09T14:37:14.2086953Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:37:14.2087626Z triton_mm_408 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=1 2025-09-09T14:37:14.2088643Z triton_mm_410 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=1 2025-09-09T14:37:14.2089657Z triton_mm_407 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=1 2025-09-09T14:37:14.2090666Z triton_mm_409 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=1 2025-09-09T14:37:14.2091669Z triton_mm_411 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=16, EVEN_K=False, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=1 2025-09-09T14:37:14.2092308Z mm 0.0379 ms 64.9% 2025-09-09T14:37:14.2092760Z SingleProcess AUTOTUNE benchmarking takes 0.0777 seconds and 0.2900 seconds precompiling for 6 choices 2025-09-09T14:37:14.2093484Z PASSED 2025-09-09T14:37:14.2094167Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_0_cpu SKIPPED 2025-09-09T14:37:14.2095176Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_1_cpu SKIPPED 2025-09-09T14:37:14.2096188Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_2_cpu SKIPPED 2025-09-09T14:37:14.2097279Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_3_cuda PASSED 2025-09-09T14:37:14.2098279Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_4_cuda PASSED 2025-09-09T14:37:14.2099355Z test/integration/test_integration.py::TestWeightOnlyInt8Quant::test_weight_only_quant_use_mixed_mm_5_cuda PASSED 2025-09-09T14:37:14.2100276Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_0_cpu SKIPPED 2025-09-09T14:37:14.2101132Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_1_cpu SKIPPED 2025-09-09T14:37:14.2101975Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_2_cpu SKIPPED 2025-09-09T14:37:14.2102778Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_3_cuda Autotune Choices Stats: 2025-09-09T14:37:14.2104100Z {"num_choices": 5, "num_triton_choices": 4, "best_kernel": "triton_mm_482", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:37:14.2105074Z AUTOTUNE int_mm(32x32, 32x32) 2025-09-09T14:37:14.2105328Z strides: [32, 1], [1, 32] 2025-09-09T14:37:14.2105567Z dtypes: torch.int8, torch.int8 2025-09-09T14:37:14.2106187Z triton_mm_482 0.0236 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:14.2107170Z triton_mm_480 0.0256 ms 92.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:14.2108133Z triton_mm_481 0.0256 ms 92.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:14.2109108Z triton_mm_483 0.0256 ms 92.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:14.2109724Z _int_mm 0.0389 ms 60.5% 2025-09-09T14:37:14.2110196Z SingleProcess AUTOTUNE benchmarking takes 0.0653 seconds and 0.1982 seconds precompiling for 5 choices 2025-09-09T14:37:14.2110711Z Autotune Choices Stats: 2025-09-09T14:37:14.2111639Z {"num_choices": 6, "num_triton_choices": 5, "best_kernel": "triton_mm_477", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.024607999250292778, "best_triton_pos": 0} 2025-09-09T14:37:14.2112610Z AUTOTUNE int_mm(32x64, 64x32) 2025-09-09T14:37:14.2112864Z strides: [64, 1], [1, 64] 2025-09-09T14:37:14.2113100Z dtypes: torch.int8, torch.int8 2025-09-09T14:37:14.2113723Z triton_mm_477 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:14.2114700Z triton_mm_476 0.0256 ms 96.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:14.2115671Z triton_mm_475 0.0256 ms 96.1% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:14.2116692Z triton_mm_478 0.0256 ms 96.1% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:14.2117653Z triton_mm_479 0.0256 ms 96.1% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:14.2118371Z _int_mm 0.0389 ms 63.2% 2025-09-09T14:37:14.2118839Z SingleProcess AUTOTUNE benchmarking takes 0.0830 seconds and 0.0002 seconds precompiling for 6 choices 2025-09-09T14:37:14.2119379Z PASSED 2025-09-09T14:37:14.2120006Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_4_cuda PASSED 2025-09-09T14:37:14.2120859Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_dqtensors_5_cuda PASSED 2025-09-09T14:37:14.2121736Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_0_cpu SKIPPED 2025-09-09T14:37:14.2122627Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_1_cpu SKIPPED 2025-09-09T14:37:23.2684003Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_2_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2685084Z File "/pytorch/ao/test/integration/test_integration.py", line 1248, in forward 2025-09-09T14:37:23.2685540Z x = self.lin1(x) 2025-09-09T14:37:23.2685729Z 2025-09-09T14:37:23.2685936Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2686609Z File "/pytorch/ao/test/integration/test_integration.py", line 1249, in forward 2025-09-09T14:37:23.2687078Z x = self.relu(x) 2025-09-09T14:37:23.2687219Z 2025-09-09T14:37:23.2687365Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2687829Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2688244Z x = self.lin2(x) 2025-09-09T14:37:23.2688374Z 2025-09-09T14:37:23.2688685Z PASSED 2025-09-09T14:37:23.2689262Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_3_cuda SKIPPED 2025-09-09T14:37:23.2690155Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_4_cuda SKIPPED 2025-09-09T14:37:23.2690984Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int4woqtensors_5_cuda Autotune Choices Stats: 2025-09-09T14:37:23.2692276Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_502", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025567999109625816, "best_triton_pos": 0} 2025-09-09T14:37:23.2693277Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:37:23.2693508Z strides: [64, 1], [1, 64] 2025-09-09T14:37:23.2693766Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:37:23.2694425Z triton_mm_502 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:23.2695442Z triton_mm_503 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:23.2696650Z triton_mm_504 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:23.2697877Z triton_mm_505 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:23.2699133Z triton_mm_506 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:23.2700160Z triton_mm_507 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:23.2701142Z triton_mm_508 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:23.2702481Z triton_mm_509 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:23.2703628Z triton_mm_510 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:23.2704613Z triton_mm_511 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:23.2705487Z SingleProcess AUTOTUNE benchmarking takes 0.1437 seconds and 0.2066 seconds precompiling for 11 choices 2025-09-09T14:37:23.2705993Z Autotune Choices Stats: 2025-09-09T14:37:23.2706946Z {"num_choices": 8, "num_triton_choices": 6, "best_kernel": "triton_mm_512", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:37:23.2707932Z AUTOTUNE addmm(32x32, 32x32, 32x32) 2025-09-09T14:37:23.2708203Z strides: [0, 1], [32, 1], [1, 32] 2025-09-09T14:37:23.2708529Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:37:23.2709223Z triton_mm_512 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:23.2710222Z triton_mm_515 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:23.2711218Z triton_mm_516 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:23.2712205Z triton_mm_513 0.0266 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:23.2713191Z triton_mm_514 0.0266 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:23.2714188Z triton_mm_517 0.0266 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:23.2714808Z addmm 0.0502 ms 51.0% 2025-09-09T14:37:23.2715043Z bias_addmm 0.0696 ms 36.8% 2025-09-09T14:37:23.2715514Z SingleProcess AUTOTUNE benchmarking takes 0.1047 seconds and 0.0002 seconds precompiling for 8 choices 2025-09-09T14:37:23.2716069Z PASSED 2025-09-09T14:37:23.2716754Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_0_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2717569Z File "/pytorch/ao/test/integration/test_integration.py", line 1248, in forward 2025-09-09T14:37:23.2717985Z x = self.lin1(x) 2025-09-09T14:37:23.2718119Z 2025-09-09T14:37:23.2718262Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2718741Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2719150Z x = self.lin2(x) 2025-09-09T14:37:23.2719287Z 2025-09-09T14:37:23.2719430Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2719900Z File "/pytorch/ao/test/integration/test_integration.py", line 1248, in forward 2025-09-09T14:37:23.2720305Z x = self.lin1(x) 2025-09-09T14:37:23.2720435Z 2025-09-09T14:37:23.2720586Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2721045Z File "/pytorch/ao/test/integration/test_integration.py", line 1249, in forward 2025-09-09T14:37:23.2721938Z x = self.relu(x) 2025-09-09T14:37:23.2722068Z 2025-09-09T14:37:23.2722209Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2722675Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2723080Z x = self.lin2(x) 2025-09-09T14:37:23.2723210Z 2025-09-09T14:37:23.2723491Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2723959Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2724387Z x = self.lin2(x) 2025-09-09T14:37:23.2724520Z 2025-09-09T14:37:23.2724643Z PASSED 2025-09-09T14:37:23.2725249Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_1_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2726052Z File "/pytorch/ao/test/integration/test_integration.py", line 1248, in forward 2025-09-09T14:37:23.2726462Z x = self.lin1(x) 2025-09-09T14:37:23.2726597Z 2025-09-09T14:37:23.2726740Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2727209Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2727617Z x = self.lin2(x) 2025-09-09T14:37:23.2727745Z 2025-09-09T14:37:23.2727895Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2728366Z File "/pytorch/ao/test/integration/test_integration.py", line 1248, in forward 2025-09-09T14:37:23.2728767Z x = self.lin1(x) 2025-09-09T14:37:23.2728903Z 2025-09-09T14:37:23.2729049Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2729560Z File "/pytorch/ao/test/integration/test_integration.py", line 1249, in forward 2025-09-09T14:37:23.2729978Z x = self.relu(x) 2025-09-09T14:37:23.2730108Z 2025-09-09T14:37:23.2730257Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2730718Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2731144Z x = self.lin2(x) 2025-09-09T14:37:23.2731273Z 2025-09-09T14:37:23.2731417Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2731882Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2732292Z x = self.lin2(x) 2025-09-09T14:37:23.2732421Z 2025-09-09T14:37:23.2740329Z PASSED 2025-09-09T14:37:23.2741015Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_2_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2741844Z File "/pytorch/ao/test/integration/test_integration.py", line 1248, in forward 2025-09-09T14:37:23.2742275Z x = self.lin1(x) 2025-09-09T14:37:23.2742413Z 2025-09-09T14:37:23.2742563Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2743045Z File "/pytorch/ao/test/integration/test_integration.py", line 1249, in forward 2025-09-09T14:37:23.2743455Z x = self.relu(x) 2025-09-09T14:37:23.2743603Z 2025-09-09T14:37:23.2743751Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2744230Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2744636Z x = self.lin2(x) 2025-09-09T14:37:23.2744777Z 2025-09-09T14:37:23.2744926Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:37:23.2745396Z File "/pytorch/ao/test/integration/test_integration.py", line 1250, in forward 2025-09-09T14:37:23.2745811Z x = self.lin2(x) 2025-09-09T14:37:23.2745944Z 2025-09-09T14:37:23.2746101Z PASSED 2025-09-09T14:37:23.2746610Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_3_cuda Autotune Choices Stats: 2025-09-09T14:37:28.0332917Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_521", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:37:28.0334449Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:37:28.0334694Z strides: [64, 1], [32, 1] 2025-09-09T14:37:28.0334961Z dtypes: torch.float32, torch.float32 2025-09-09T14:37:28.0335788Z triton_mm_521 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:28.0336796Z triton_mm_522 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0337785Z triton_mm_524 0.0245 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:28.0338751Z triton_mm_519 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0339724Z triton_mm_520 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:28.0340702Z triton_mm_525 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:28.0341689Z triton_mm_527 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:28.0342687Z triton_mm_518 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:28.0343648Z triton_mm_523 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0344623Z triton_mm_526 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:28.0345480Z SingleProcess AUTOTUNE benchmarking takes 0.1372 seconds and 0.2961 seconds precompiling for 11 choices 2025-09-09T14:37:28.0345982Z Autotune Choices Stats: 2025-09-09T14:37:28.0346923Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_529", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:37:28.0347881Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:37:28.0348110Z strides: [32, 1], [32, 1] 2025-09-09T14:37:28.0348356Z dtypes: torch.float32, torch.float32 2025-09-09T14:37:28.0348993Z triton_mm_529 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0349983Z triton_mm_531 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0350955Z triton_mm_528 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:28.0351922Z triton_mm_530 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:28.0352893Z triton_mm_532 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:28.0353952Z triton_mm_533 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:28.0354555Z mm 0.0379 ms 64.9% 2025-09-09T14:37:28.0355079Z SingleProcess AUTOTUNE benchmarking takes 0.0871 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:37:28.0355786Z PASSED 2025-09-09T14:37:28.0356420Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_4_cuda Autotune Choices Stats: 2025-09-09T14:37:28.0357689Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_534", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:37:28.0358645Z AUTOTUNE mm(32x64, 64x32) 2025-09-09T14:37:28.0358876Z strides: [64, 1], [32, 1] 2025-09-09T14:37:28.0359126Z dtypes: torch.float16, torch.float16 2025-09-09T14:37:28.0359767Z triton_mm_534 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:28.0360757Z triton_mm_542 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:28.0361768Z triton_mm_537 0.0246 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:28.0362750Z triton_mm_541 0.0246 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:28.0364273Z triton_mm_540 0.0255 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:28.0365636Z triton_mm_535 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0366624Z triton_mm_536 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:28.0367598Z triton_mm_538 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0368580Z triton_mm_539 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0369558Z triton_mm_543 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:28.0370423Z SingleProcess AUTOTUNE benchmarking takes 0.1401 seconds and 0.1804 seconds precompiling for 11 choices 2025-09-09T14:37:28.0370941Z Autotune Choices Stats: 2025-09-09T14:37:28.0371890Z {"num_choices": 7, "num_triton_choices": 6, "best_kernel": "triton_mm_547", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:37:28.0372850Z AUTOTUNE mm(32x32, 32x32) 2025-09-09T14:37:28.0373089Z strides: [32, 1], [32, 1] 2025-09-09T14:37:28.0373333Z dtypes: torch.float16, torch.float16 2025-09-09T14:37:28.0373971Z triton_mm_547 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0375153Z triton_mm_546 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:28.0376236Z triton_mm_544 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:28.0377220Z triton_mm_545 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:28.0378194Z triton_mm_548 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:28.0379161Z triton_mm_549 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:37:28.0379778Z mm 0.0389 ms 60.5% 2025-09-09T14:37:28.0380223Z SingleProcess AUTOTUNE benchmarking takes 0.0888 seconds and 0.0002 seconds precompiling for 7 choices 2025-09-09T14:37:28.0380788Z PASSED 2025-09-09T14:37:28.0381373Z test/integration/test_integration.py::TestSaveLoadMeta::test_save_load_int8woqtensors_5_cuda PASSED 2025-09-09T14:37:54.2423525Z test/integration/test_integration.py::TorchCompileUnitTest::test_fullgraph PASSED 2025-09-09T14:37:54.2427364Z test/integration/test_integration.py::UtilsUnitTest::test_shape_logger PASSED 2025-09-09T14:37:54.2428356Z test/integration/test_integration.py::SmoothquantIntegrationTest::test_non_dynamically_quantizable_linear SKIPPED 2025-09-09T14:37:54.2429166Z test/integration/test_integration.py::SmoothquantIntegrationTest::test_on_dummy_distilbert 2025-09-09T14:37:54.2429698Z tokenizer_config.json: 0% 0.00/48.0 [00:00>time: 0.011ms for , to_beat: infms 2025-09-09T14:37:54.2459890Z Autotune Choices Stats: 2025-09-09T14:37:54.2460836Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_585", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:37:54.2461902Z AUTOTUNE mm(32x128, 128x128) 2025-09-09T14:37:54.2462142Z strides: [128, 1], [128, 1] 2025-09-09T14:37:54.2462404Z dtypes: torch.float32, torch.float32 2025-09-09T14:37:54.2463122Z triton_mm_585 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:37:54.2464273Z triton_mm_583 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:37:54.2465356Z triton_mm_586 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:54.2466566Z triton_mm_591 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:54.2467801Z triton_mm_592 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:54.2469027Z triton_mm_596 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:37:54.2470245Z triton_mm_584 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:54.2471473Z triton_mm_587 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:37:54.2472685Z triton_mm_588 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:54.2473903Z triton_mm_589 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:37:54.2474962Z SingleProcess AUTOTUNE benchmarking takes 0.2095 seconds and 1.8043 seconds precompiling for 18 choices 2025-09-09T14:37:54.2475975Z >>time: 0.016ms for , to_beat: 0.011ms 2025-09-09T14:37:54.2477029Z >>time: 0.014ms for , to_beat: 0.011ms 2025-09-09T14:37:54.2477633Z Autotune Choices Stats: 2025-09-09T14:37:54.2478557Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_600", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:37:54.2479520Z AUTOTUNE int_mm(32x128, 128x128) 2025-09-09T14:37:54.2479772Z strides: [128, 1], [1, 128] 2025-09-09T14:37:54.2480020Z dtypes: torch.int8, torch.int8 2025-09-09T14:37:54.2480628Z triton_mm_600 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6479092Z triton_mm_601 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:23.6480113Z triton_mm_602 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:23.6481136Z triton_mm_603 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:38:23.6484148Z triton_mm_604 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:38:23.6485283Z triton_mm_605 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:23.6486263Z triton_mm_606 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:38:23.6487233Z triton_mm_607 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:38:23.6488204Z triton_mm_608 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:38:23.6489182Z triton_mm_609 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:38:23.6490057Z SingleProcess AUTOTUNE benchmarking takes 0.1369 seconds and 0.3026 seconds precompiling for 11 choices 2025-09-09T14:38:23.6490934Z >>time: 0.014ms for matmul, to_beat: 0.011ms 2025-09-09T14:38:23.6491712Z best_cls= 2025-09-09T14:38:23.6492048Z 2025-09-09T14:38:23.6492354Z PASSED 2025-09-09T14:38:23.6492915Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_12_cuda SKIPPED 2025-09-09T14:38:23.6493737Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_13_cuda SKIPPED 2025-09-09T14:38:23.6494614Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_14_cuda activation_shapes: torch.Size([32, 128]), times_seen: 2 2025-09-09T14:38:23.6495405Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:38:23.6495830Z Autotune Choices Stats: 2025-09-09T14:38:23.6496796Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_621", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:38:23.6497847Z AUTOTUNE addmm(32x128, 32x128, 128x128) 2025-09-09T14:38:23.6498130Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:38:23.6498441Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:38:23.6499128Z triton_mm_621 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:23.6500135Z triton_mm_613 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:23.6501114Z triton_mm_617 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:38:23.6502092Z triton_mm_610 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:38:23.6503073Z triton_mm_611 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6504050Z triton_mm_612 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:38:23.6505129Z triton_mm_614 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:23.6506187Z triton_mm_615 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6507159Z triton_mm_616 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6508188Z triton_mm_618 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:23.6509050Z SingleProcess AUTOTUNE benchmarking takes 0.2382 seconds and 0.4178 seconds precompiling for 19 choices 2025-09-09T14:38:23.6509782Z >>time: 0.009ms for , to_beat: infms 2025-09-09T14:38:23.6510288Z Autotune Choices Stats: 2025-09-09T14:38:23.6511233Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_643", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8", "best_time": 0.0225600004196167, "best_triton_pos": 0} 2025-09-09T14:38:23.6512204Z AUTOTUNE mm(32x128, 128x128) 2025-09-09T14:38:23.6512451Z strides: [128, 1], [128, 1] 2025-09-09T14:38:23.6512702Z dtypes: torch.float16, torch.float16 2025-09-09T14:38:23.6513349Z triton_mm_643 0.0226 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:38:23.6514344Z triton_mm_636 0.0236 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:23.6515340Z triton_mm_638 0.0236 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:23.6516428Z triton_mm_631 0.0236 ms 95.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:23.6517908Z triton_mm_627 0.0246 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:38:23.6519268Z triton_mm_628 0.0246 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6520250Z triton_mm_629 0.0246 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:38:23.6521222Z triton_mm_630 0.0246 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:23.6522195Z triton_mm_632 0.0246 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6523166Z triton_mm_633 0.0246 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:23.6524021Z SingleProcess AUTOTUNE benchmarking takes 0.2186 seconds and 0.3892 seconds precompiling for 18 choices 2025-09-09T14:38:23.6524851Z >>time: 0.014ms for , to_beat: 0.009ms 2025-09-09T14:38:23.6525870Z >>time: 0.014ms for , to_beat: 0.009ms 2025-09-09T14:38:23.6526888Z >>time: 0.008ms for matmul, to_beat: 0.009ms 2025-09-09T14:38:23.6527839Z >>time: 0.009ms for , to_beat: 0.012ms 2025-09-09T14:38:23.6528838Z >>time: 0.009ms for interpolated, breakeven constant: 0.58 2025-09-09T14:38:23.6529673Z best_cls= 2025-09-09T14:38:23.6530007Z 2025-09-09T14:38:23.6530140Z PASSED 2025-09-09T14:38:23.6530668Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_15_cuda SKIPPED 2025-09-09T14:38:23.6531494Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_16_cuda SKIPPED 2025-09-09T14:38:48.8356568Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_compile_17_cuda activation_shapes: torch.Size([32, 128]), times_seen: 2 2025-09-09T14:38:48.8359283Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:38:48.8359774Z Autotune Choices Stats: 2025-09-09T14:38:48.8360734Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_673", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:38:48.8361753Z AUTOTUNE addmm(32x128, 32x128, 128x128) 2025-09-09T14:38:48.8362038Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:38:48.8362357Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:38:48.8363084Z triton_mm_673 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:48.8364402Z triton_mm_680 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:38:48.8365410Z triton_mm_665 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8366398Z triton_mm_666 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:38:48.8367380Z triton_mm_668 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:48.8368368Z triton_mm_669 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8369359Z triton_mm_672 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:48.8370343Z triton_mm_674 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:38:48.8371338Z triton_mm_676 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:38:48.8372330Z triton_mm_677 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:48.8373549Z SingleProcess AUTOTUNE benchmarking takes 0.2451 seconds and 0.3944 seconds precompiling for 19 choices 2025-09-09T14:38:48.8374294Z >>time: 0.009ms for , to_beat: infms 2025-09-09T14:38:48.8374806Z Autotune Choices Stats: 2025-09-09T14:38:48.8375905Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_692", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:38:48.8376889Z AUTOTUNE mm(32x128, 128x128) 2025-09-09T14:38:48.8377136Z strides: [128, 1], [128, 1] 2025-09-09T14:38:48.8377396Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:38:48.8378044Z triton_mm_692 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:48.8379049Z triton_mm_682 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8380040Z triton_mm_687 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8381012Z triton_mm_688 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:38:48.8381991Z triton_mm_690 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:48.8382981Z triton_mm_691 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:38:48.8384003Z triton_mm_694 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:48.8384983Z triton_mm_686 0.0246 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8385974Z triton_mm_693 0.0246 ms 95.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:38:48.8386956Z triton_mm_681 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:38:48.8387816Z SingleProcess AUTOTUNE benchmarking takes 0.2259 seconds and 0.3762 seconds precompiling for 18 choices 2025-09-09T14:38:48.8388656Z >>time: 0.015ms for , to_beat: 0.009ms 2025-09-09T14:38:48.8389571Z >>time: 0.017ms for , to_beat: 0.009ms 2025-09-09T14:38:48.8390525Z >>time: 0.014ms for matmul, to_beat: 0.009ms 2025-09-09T14:38:48.8391287Z best_cls= 2025-09-09T14:38:48.8391627Z 2025-09-09T14:38:48.8391931Z PASSED 2025-09-09T14:38:48.8392530Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_0_cpu SKIPPED 2025-09-09T14:38:48.8393408Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_1_cpu SKIPPED 2025-09-09T14:38:48.8394269Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_2_cpu SKIPPED 2025-09-09T14:38:48.8395261Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_3_cuda activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:38:48.8396071Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:38:48.8396685Z Autotune Choices Stats: 2025-09-09T14:38:48.8397638Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_712", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.023584000766277313, "best_triton_pos": 0} 2025-09-09T14:38:48.8398628Z AUTOTUNE addmm(16x128, 16x128, 128x128) 2025-09-09T14:38:48.8398898Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:38:48.8399206Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:38:48.8399899Z triton_mm_712 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:38:48.8400900Z triton_mm_715 0.0256 ms 92.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:38:48.8401895Z triton_mm_708 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:38:48.8402877Z triton_mm_709 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:38:48.8403857Z triton_mm_710 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:38:48.8404839Z triton_mm_711 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:38:48.8405821Z triton_mm_713 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8406802Z triton_mm_714 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:38:48.8407785Z triton_mm_721 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0029909Z triton_mm_722 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:06.0030856Z SingleProcess AUTOTUNE benchmarking takes 0.2409 seconds and 1.9196 seconds precompiling for 19 choices 2025-09-09T14:39:06.0031591Z >>time: 0.008ms for , to_beat: infms 2025-09-09T14:39:06.0032101Z Autotune Choices Stats: 2025-09-09T14:39:06.0033075Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_726", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:39:06.0034074Z AUTOTUNE mm(16x128, 128x128) 2025-09-09T14:39:06.0034522Z strides: [128, 1], [128, 1] 2025-09-09T14:39:06.0034778Z dtypes: torch.float32, torch.float32 2025-09-09T14:39:06.0035437Z triton_mm_726 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:39:06.0036914Z triton_mm_738 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0038069Z triton_mm_725 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:39:06.0039057Z triton_mm_727 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:06.0040034Z triton_mm_730 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:06.0041024Z triton_mm_731 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:06.0042011Z triton_mm_732 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0042993Z triton_mm_733 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:06.0043974Z triton_mm_734 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0044959Z triton_mm_735 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:39:06.0045819Z SingleProcess AUTOTUNE benchmarking takes 0.2166 seconds and 1.0438 seconds precompiling for 18 choices 2025-09-09T14:39:06.0046649Z >>time: 0.015ms for , to_beat: 0.008ms 2025-09-09T14:39:06.0047566Z >>time: 0.014ms for , to_beat: 0.008ms 2025-09-09T14:39:06.0048163Z Autotune Choices Stats: 2025-09-09T14:39:06.0049101Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_746", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:39:06.0050068Z AUTOTUNE int_mm(16x128, 128x128) 2025-09-09T14:39:06.0050318Z strides: [128, 1], [1, 128] 2025-09-09T14:39:06.0050569Z dtypes: torch.int8, torch.int8 2025-09-09T14:39:06.0051177Z triton_mm_746 0.0236 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:06.0052142Z triton_mm_742 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:06.0053105Z triton_mm_743 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0054058Z triton_mm_744 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0055017Z triton_mm_745 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:39:06.0055972Z triton_mm_747 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:06.0057011Z triton_mm_748 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:06.0058042Z triton_mm_749 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:39:06.0058998Z triton_mm_750 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0059956Z triton_mm_751 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:39:06.0060812Z SingleProcess AUTOTUNE benchmarking takes 0.1290 seconds and 0.2856 seconds precompiling for 11 choices 2025-09-09T14:39:06.0061671Z >>time: 0.015ms for matmul, to_beat: 0.008ms 2025-09-09T14:39:06.0062446Z best_cls= 2025-09-09T14:39:06.0062781Z 2025-09-09T14:39:06.0063091Z PASSED 2025-09-09T14:39:06.0063952Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_4_cuda activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:39:06.0064835Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:39:06.0065256Z Autotune Choices Stats: 2025-09-09T14:39:06.0066199Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_766", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:39:06.0067184Z AUTOTUNE addmm(16x128, 16x128, 128x128) 2025-09-09T14:39:06.0067453Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:39:06.0067766Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:39:06.0068452Z triton_mm_766 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:06.0069452Z triton_mm_763 0.0236 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:06.0070445Z triton_mm_767 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:39:06.0071423Z triton_mm_758 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:06.0072411Z triton_mm_752 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:39:06.0073401Z triton_mm_753 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:39:06.0074378Z triton_mm_754 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:06.0075361Z triton_mm_755 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:06.0076399Z triton_mm_756 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:06.0078597Z triton_mm_757 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:06.0079573Z SingleProcess AUTOTUNE benchmarking takes 0.2436 seconds and 0.3803 seconds precompiling for 19 choices 2025-09-09T14:39:26.9708864Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:39:26.9709402Z Autotune Choices Stats: 2025-09-09T14:39:26.9710384Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_781", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:39:26.9711393Z AUTOTUNE mm(16x128, 128x128) 2025-09-09T14:39:26.9711641Z strides: [128, 1], [128, 1] 2025-09-09T14:39:26.9711926Z dtypes: torch.float16, torch.float16 2025-09-09T14:39:26.9712582Z triton_mm_781 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:26.9713607Z triton_mm_782 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:26.9714599Z triton_mm_771 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:26.9715574Z triton_mm_772 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:26.9716668Z triton_mm_773 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:26.9717660Z triton_mm_774 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:26.9718639Z triton_mm_776 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:26.9719647Z triton_mm_777 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:26.9720645Z triton_mm_779 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:39:26.9721634Z triton_mm_780 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:26.9722513Z SingleProcess AUTOTUNE benchmarking takes 0.2254 seconds and 0.3252 seconds precompiling for 18 choices 2025-09-09T14:39:26.9723349Z >>time: 0.014ms for , to_beat: 0.014ms 2025-09-09T14:39:26.9724267Z >>time: 0.015ms for , to_beat: 0.014ms 2025-09-09T14:39:26.9725219Z >>time: 0.015ms for matmul, to_beat: 0.014ms 2025-09-09T14:39:26.9726063Z best_cls= 2025-09-09T14:39:26.9726490Z 2025-09-09T14:39:26.9726790Z PASSED 2025-09-09T14:39:26.9727542Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_double_access_5_cuda activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:39:26.9728741Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:39:26.9729175Z Autotune Choices Stats: 2025-09-09T14:39:26.9731798Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_797", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:39:26.9732815Z AUTOTUNE addmm(16x128, 16x128, 128x128) 2025-09-09T14:39:26.9733096Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:39:26.9733414Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:39:26.9734124Z triton_mm_797 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:39:26.9735122Z triton_mm_810 0.0246 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:26.9736133Z triton_mm_800 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:26.9737129Z triton_mm_805 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:26.9738170Z triton_mm_808 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:26.9739162Z triton_mm_803 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:26.9740146Z triton_mm_802 0.0266 ms 92.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:26.9741125Z triton_mm_796 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:39:26.9742105Z triton_mm_798 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:26.9743080Z triton_mm_799 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:26.9743941Z SingleProcess AUTOTUNE benchmarking takes 0.2449 seconds and 0.3612 seconds precompiling for 19 choices 2025-09-09T14:39:26.9744675Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:39:26.9745186Z Autotune Choices Stats: 2025-09-09T14:39:26.9746131Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_816", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:39:26.9747106Z AUTOTUNE mm(16x128, 128x128) 2025-09-09T14:39:26.9747346Z strides: [128, 1], [128, 1] 2025-09-09T14:39:26.9747610Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:39:26.9748255Z triton_mm_816 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:26.9749253Z triton_mm_829 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:39:26.9750334Z triton_mm_813 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:39:26.9751392Z triton_mm_814 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:39:26.9752378Z triton_mm_815 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:26.9753362Z triton_mm_817 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:26.9754340Z triton_mm_818 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:26.9755325Z triton_mm_819 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:26.9756407Z triton_mm_820 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:26.9757397Z triton_mm_821 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:26.9758321Z SingleProcess AUTOTUNE benchmarking takes 0.2229 seconds and 0.3378 seconds precompiling for 18 choices 2025-09-09T14:39:26.9759159Z >>time: 0.014ms for , to_beat: 0.014ms 2025-09-09T14:39:44.8749532Z >>time: 0.014ms for , to_beat: 0.014ms 2025-09-09T14:39:44.8750945Z >>time: 0.015ms for matmul, to_beat: 0.014ms 2025-09-09T14:39:44.8751848Z best_cls= 2025-09-09T14:39:44.8752289Z 2025-09-09T14:39:44.8752611Z PASSED 2025-09-09T14:39:44.8753166Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_0_cpu SKIPPED 2025-09-09T14:39:44.8753967Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_1_cpu SKIPPED 2025-09-09T14:39:44.8754759Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_2_cpu SKIPPED 2025-09-09T14:39:44.8755546Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_3_cuda SKIPPED 2025-09-09T14:39:44.8756429Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_4_cuda SKIPPED 2025-09-09T14:39:44.8757234Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_float8_5_cuda SKIPPED 2025-09-09T14:39:44.8758065Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_hp_float activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:39:44.8758838Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:39:44.8759494Z >>time: 0.009ms for , to_beat: infms 2025-09-09T14:39:44.8760136Z best_cls= 2025-09-09T14:39:44.8760469Z 2025-09-09T14:39:44.8760623Z activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:39:44.8761100Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:39:44.8761753Z >>time: 0.012ms for , to_beat: infms 2025-09-09T14:39:44.8762699Z best_cls= 2025-09-09T14:39:44.8763027Z 2025-09-09T14:39:44.8763177Z activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:39:44.8763964Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:39:44.8764614Z >>time: 0.012ms for , to_beat: infms 2025-09-09T14:39:44.8765251Z best_cls= 2025-09-09T14:39:44.8765575Z 2025-09-09T14:39:44.8765703Z PASSED 2025-09-09T14:39:44.8766212Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_0_cpu SKIPPED 2025-09-09T14:39:44.8767002Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_1_cpu SKIPPED 2025-09-09T14:39:44.8767778Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_2_cpu SKIPPED 2025-09-09T14:39:44.8768569Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_3_cuda SKIPPED 2025-09-09T14:39:44.8769408Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_4_cuda SKIPPED 2025-09-09T14:39:44.8770209Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_int4wo_5_cuda SKIPPED 2025-09-09T14:39:44.8771001Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_00_cpu SKIPPED 2025-09-09T14:39:44.8771784Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_01_cpu SKIPPED 2025-09-09T14:39:44.8772574Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_02_cpu SKIPPED 2025-09-09T14:39:44.8773354Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_03_cpu SKIPPED 2025-09-09T14:39:44.8774144Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_04_cpu SKIPPED 2025-09-09T14:39:44.8774945Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_05_cpu SKIPPED 2025-09-09T14:39:44.8775727Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_06_cpu SKIPPED 2025-09-09T14:39:44.8776520Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_07_cpu SKIPPED 2025-09-09T14:39:44.8777303Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_08_cpu SKIPPED 2025-09-09T14:39:44.8778101Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_09_cuda SKIPPED 2025-09-09T14:39:44.8778926Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_10_cuda SKIPPED 2025-09-09T14:39:44.8779746Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_11_cuda PASSED 2025-09-09T14:39:44.8780544Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_12_cuda SKIPPED 2025-09-09T14:39:44.8781343Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_13_cuda SKIPPED 2025-09-09T14:39:44.8782132Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_14_cuda PASSED 2025-09-09T14:39:44.8782932Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_15_cuda SKIPPED 2025-09-09T14:39:44.8783725Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_16_cuda SKIPPED 2025-09-09T14:39:44.8784519Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_kwargs_17_cuda PASSED 2025-09-09T14:39:44.8785302Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_0_cpu SKIPPED 2025-09-09T14:39:44.8786089Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_1_cpu SKIPPED 2025-09-09T14:39:44.8786871Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_2_cpu SKIPPED 2025-09-09T14:39:44.8787800Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_3_cuda PASSED 2025-09-09T14:39:44.8788581Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_4_cuda PASSED 2025-09-09T14:39:44.8789513Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_manual_5_cuda PASSED 2025-09-09T14:39:44.8790289Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_0_cpu SKIPPED 2025-09-09T14:39:44.8791049Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_1_cpu SKIPPED 2025-09-09T14:39:44.8791800Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_2_cpu SKIPPED 2025-09-09T14:39:44.8792615Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_3_cuda activation_shapes: torch.Size([1, 4096]), times_seen: 1 2025-09-09T14:39:44.8793385Z weight_shape: torch.Size([4096, 4096]), dtype: torch.float32, bias_shape: torch.Size([4096]) 2025-09-09T14:39:44.8793840Z Autotune Choices Stats: 2025-09-09T14:39:44.8795059Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "bias_addmm", "best_time": 0.14336000382900238, "best_triton_pos": 2, "best_triton_time": 0.18636800348758698, "best_triton_kernel": "triton_mm_847", "best_triton_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4"} 2025-09-09T14:39:44.8796378Z AUTOTUNE addmm(1x4096, 1x4096, 4096x4096) 2025-09-09T14:39:44.8796663Z strides: [0, 1], [4096, 1], [1, 4096] 2025-09-09T14:39:44.8796964Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:39:44.8797284Z bias_addmm 0.1434 ms 100.0% 2025-09-09T14:39:44.8797523Z addmm 0.1495 ms 95.9% 2025-09-09T14:39:44.8798122Z triton_mm_847 0.1864 ms 76.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:44.8799151Z triton_mm_842 0.1894 ms 75.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:39:44.8800145Z triton_mm_854 0.1905 ms 75.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:39:44.8801127Z triton_mm_853 0.1915 ms 74.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:39:44.8802110Z triton_mm_843 0.2048 ms 70.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:44.8803088Z triton_mm_844 0.2068 ms 69.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:39:44.8804083Z triton_mm_841 0.2243 ms 63.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:39:44.8805078Z triton_mm_846 0.2560 ms 56.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:39:44.8805936Z SingleProcess AUTOTUNE benchmarking takes 0.4810 seconds and 1.3963 seconds precompiling for 19 choices 2025-09-09T14:39:44.8806670Z >>time: 0.145ms for , to_beat: infms 2025-09-09T14:39:44.8807497Z >>time: 0.038ms for , to_beat: 0.145ms 2025-09-09T14:39:44.8808409Z >>time: 0.038ms for , to_beat: 0.038ms 2025-09-09T14:39:44.8809153Z Autotune Choices Stats: 2025-09-09T14:40:15.5611973Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_866", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=256, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8", "best_time": 0.04915200173854828, "best_triton_pos": 0} 2025-09-09T14:40:15.5613045Z AUTOTUNE int_mm(1x4096, 4096x4096) 2025-09-09T14:40:15.5613314Z strides: [4096, 1], [1, 4096] 2025-09-09T14:40:15.5613578Z dtypes: torch.int8, torch.int8 2025-09-09T14:40:15.5614197Z triton_mm_866 0.0492 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=256, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:40:15.5615189Z triton_mm_867 0.0502 ms 98.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:40:15.5616160Z triton_mm_865 0.0532 ms 92.3% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5617115Z triton_mm_862 0.0543 ms 90.6% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:40:15.5618059Z triton_mm_863 0.0553 ms 88.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:15.5618996Z triton_mm_861 0.0604 ms 81.4% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:40:15.5619941Z triton_mm_859 0.0737 ms 66.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5620899Z triton_mm_858 0.0778 ms 63.2% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5621853Z triton_mm_857 0.1280 ms 38.4% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:15.5622803Z triton_mm_860 0.1290 ms 38.1% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:40:15.5623656Z SingleProcess AUTOTUNE benchmarking takes 0.1706 seconds and 0.4489 seconds precompiling for 12 choices 2025-09-09T14:40:15.5624513Z >>time: 0.050ms for matmul, to_beat: 0.038ms 2025-09-09T14:40:15.5625365Z best_cls= 2025-09-09T14:40:15.5625793Z 2025-09-09T14:40:15.5626105Z PASSED 2025-09-09T14:40:15.5626688Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_4_cuda activation_shapes: torch.Size([1, 4096]), times_seen: 1 2025-09-09T14:40:15.5627473Z weight_shape: torch.Size([4096, 4096]), dtype: torch.float16, bias_shape: torch.Size([4096]) 2025-09-09T14:40:15.5627909Z Autotune Choices Stats: 2025-09-09T14:40:15.5628907Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_876", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.08089599758386612, "best_triton_pos": 0} 2025-09-09T14:40:15.5629889Z AUTOTUNE addmm(1x4096, 1x4096, 4096x4096) 2025-09-09T14:40:15.5630167Z strides: [0, 1], [4096, 1], [1, 4096] 2025-09-09T14:40:15.5630484Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:40:15.5631440Z triton_mm_876 0.0809 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:15.5632522Z triton_mm_884 0.0829 ms 97.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:15.5633511Z triton_mm_879 0.0850 ms 95.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5634483Z triton_mm_870 0.0881 ms 91.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:15.5635475Z triton_mm_872 0.0881 ms 91.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:40:15.5636524Z triton_mm_875 0.0881 ms 91.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5637146Z addmm 0.0891 ms 90.8% 2025-09-09T14:40:15.5637736Z triton_mm_882 0.0922 ms 87.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:40:15.5638364Z bias_addmm 0.0932 ms 86.8% 2025-09-09T14:40:15.5639029Z triton_mm_878 0.0963 ms 84.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:40:15.5639890Z SingleProcess AUTOTUNE benchmarking takes 0.3422 seconds and 0.5658 seconds precompiling for 19 choices 2025-09-09T14:40:15.5640630Z >>time: 0.083ms for , to_beat: infms 2025-09-09T14:40:15.5641451Z >>time: 0.038ms for , to_beat: 0.083ms 2025-09-09T14:40:15.5642381Z >>time: 0.038ms for , to_beat: 0.038ms 2025-09-09T14:40:15.5643334Z >>time: 0.050ms for matmul, to_beat: 0.038ms 2025-09-09T14:40:15.5644184Z best_cls= 2025-09-09T14:40:15.5644616Z 2025-09-09T14:40:15.5644741Z PASSED 2025-09-09T14:40:15.5645295Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_mha_5_cuda activation_shapes: torch.Size([1, 4096]), times_seen: 1 2025-09-09T14:40:15.5646080Z weight_shape: torch.Size([4096, 4096]), dtype: torch.bfloat16, bias_shape: torch.Size([4096]) 2025-09-09T14:40:15.5646527Z Autotune Choices Stats: 2025-09-09T14:40:15.5647460Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_904", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.0798719972372055, "best_triton_pos": 0} 2025-09-09T14:40:15.5648456Z AUTOTUNE addmm(1x4096, 1x4096, 4096x4096) 2025-09-09T14:40:15.5648776Z strides: [0, 1], [4096, 1], [1, 4096] 2025-09-09T14:40:15.5649135Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:40:15.5649844Z triton_mm_904 0.0799 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:15.5650833Z triton_mm_912 0.0840 ms 95.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:15.5651816Z triton_mm_907 0.0850 ms 94.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5652885Z triton_mm_903 0.0870 ms 91.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:15.5653575Z addmm 0.0881 ms 90.7% 2025-09-09T14:40:15.5654161Z triton_mm_898 0.0881 ms 90.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:15.5655131Z triton_mm_900 0.0881 ms 90.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:40:15.5655756Z bias_addmm 0.0932 ms 85.7% 2025-09-09T14:40:15.5656356Z triton_mm_910 0.0932 ms 85.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:40:15.5657332Z triton_mm_906 0.0963 ms 83.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:40:15.5658201Z SingleProcess AUTOTUNE benchmarking takes 0.3437 seconds and 0.5643 seconds precompiling for 19 choices 2025-09-09T14:40:15.5658933Z >>time: 0.082ms for , to_beat: infms 2025-09-09T14:40:15.5659752Z >>time: 0.038ms for , to_beat: 0.082ms 2025-09-09T14:40:15.5660667Z >>time: 0.038ms for , to_beat: 0.038ms 2025-09-09T14:40:15.5661605Z >>time: 0.050ms for matmul, to_beat: 0.038ms 2025-09-09T14:40:20.5892501Z best_cls= 2025-09-09T14:40:20.5892999Z 2025-09-09T14:40:20.5893314Z PASSED 2025-09-09T14:40:20.5893923Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_min_sqnr_0_cuda activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:40:20.5894729Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:40:20.5895161Z Autotune Choices Stats: 2025-09-09T14:40:20.5896109Z {"num_choices": 21, "num_triton_choices": 19, "best_kernel": "triton_mm_928", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:40:20.5897111Z AUTOTUNE addmm(128x128, 128x128, 128x128) 2025-09-09T14:40:20.5897386Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:40:20.5897696Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:40:20.5898382Z triton_mm_928 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:20.5899379Z triton_mm_930 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:20.5900363Z triton_mm_933 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:20.5901337Z triton_mm_924 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:40:20.5902314Z triton_mm_925 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:20.5903567Z triton_mm_926 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:20.5904674Z triton_mm_927 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:20.5905640Z triton_mm_929 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:20.5906601Z triton_mm_931 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:40:20.5907571Z triton_mm_932 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:20.5908443Z SingleProcess AUTOTUNE benchmarking takes 0.2372 seconds and 0.9279 seconds precompiling for 21 choices 2025-09-09T14:40:20.5909180Z >>time: 0.009ms for , to_beat: infms 2025-09-09T14:40:20.5910219Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 53.82148742675781 2025-09-09T14:40:20.5911567Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 54.006710052490234 2025-09-09T14:40:20.5912956Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 47.44115447998047 2025-09-09T14:40:20.5913896Z best_cls= 2025-09-09T14:40:20.5914234Z 2025-09-09T14:40:20.5914374Z PASSED 2025-09-09T14:40:20.5914950Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_min_sqnr_1_cuda activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:40:20.5915738Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:40:20.5916258Z Autotune Choices Stats: 2025-09-09T14:40:20.5917193Z {"num_choices": 21, "num_triton_choices": 19, "best_kernel": "triton_mm_950", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:40:20.5918170Z AUTOTUNE addmm(128x128, 128x128, 128x128) 2025-09-09T14:40:20.5918444Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:40:20.5918748Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:40:20.5919421Z triton_mm_950 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:40:20.5920426Z triton_mm_955 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:40:20.5921431Z triton_mm_959 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=128, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:20.5922427Z triton_mm_960 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=128, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:20.5923413Z triton_mm_943 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:40:20.5924518Z triton_mm_944 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:20.5925565Z triton_mm_945 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:20.5926536Z triton_mm_946 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:20.5927507Z triton_mm_947 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:20.5928473Z triton_mm_948 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:20.5929331Z SingleProcess AUTOTUNE benchmarking takes 0.2560 seconds and 0.7406 seconds precompiling for 21 choices 2025-09-09T14:40:20.5930064Z >>time: 0.012ms for , to_beat: infms 2025-09-09T14:40:20.5931063Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 54.25 2025-09-09T14:40:20.5932346Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 54.40625 2025-09-09T14:40:20.5933633Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 48.0625 2025-09-09T14:40:20.5934551Z best_cls= 2025-09-09T14:40:20.5934882Z 2025-09-09T14:40:20.5935014Z PASSED 2025-09-09T14:40:20.5935592Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_min_sqnr_2_cuda activation_shapes: torch.Size([128, 128]), times_seen: 1 2025-09-09T14:40:20.5936383Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:40:20.5936809Z Autotune Choices Stats: 2025-09-09T14:40:20.5937765Z {"num_choices": 21, "num_triton_choices": 19, "best_kernel": "triton_mm_974", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:40:20.5938754Z AUTOTUNE addmm(128x128, 128x128, 128x128) 2025-09-09T14:40:20.5939029Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:40:20.5939350Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:40:20.5940053Z triton_mm_974 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:40:20.5941062Z triton_mm_979 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=128, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:20.5942101Z triton_mm_962 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:40:20.5943068Z triton_mm_963 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6680164Z triton_mm_964 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:38.6681256Z triton_mm_965 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:38.6682608Z triton_mm_966 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:38.6683792Z triton_mm_967 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6684792Z triton_mm_968 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6685755Z triton_mm_969 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=True, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:40:38.6686629Z SingleProcess AUTOTUNE benchmarking takes 0.2553 seconds and 0.7201 seconds precompiling for 21 choices 2025-09-09T14:40:38.6687370Z >>time: 0.009ms for , to_beat: infms 2025-09-09T14:40:38.6688388Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 49.0 2025-09-09T14:40:38.6689661Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 48.75 2025-09-09T14:40:38.6690924Z skipping q_cls: because the sqnr is too small, minimum expected sqnr: 60, got 46.0 2025-09-09T14:40:38.6691839Z best_cls= 2025-09-09T14:40:38.6692175Z 2025-09-09T14:40:38.6692482Z PASSED 2025-09-09T14:40:38.6692986Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_00_cpu (m, k, n): (16, 128, 128) 2025-09-09T14:40:38.6693561Z SKIPPED 2025-09-09T14:40:38.6694033Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_01_cpu (m, k, n): (64, 128, 128) 2025-09-09T14:40:38.6694598Z SKIPPED 2025-09-09T14:40:38.6695071Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_02_cpu (m, k, n): (16, 128, 256) 2025-09-09T14:40:38.6695627Z SKIPPED 2025-09-09T14:40:38.6696099Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_03_cpu (m, k, n): (16, 256, 128) 2025-09-09T14:40:38.6696645Z SKIPPED 2025-09-09T14:40:38.6697114Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_04_cpu (m, k, n): (64, 256, 128) 2025-09-09T14:40:38.6697660Z SKIPPED 2025-09-09T14:40:38.6698128Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_05_cpu (m, k, n): (16, 128, 128) 2025-09-09T14:40:38.6698674Z SKIPPED 2025-09-09T14:40:38.6699147Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_06_cpu (m, k, n): (64, 128, 128) 2025-09-09T14:40:38.6699703Z SKIPPED 2025-09-09T14:40:38.6700172Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_07_cpu (m, k, n): (16, 128, 256) 2025-09-09T14:40:38.6700729Z SKIPPED 2025-09-09T14:40:38.6701194Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_08_cpu (m, k, n): (16, 256, 128) 2025-09-09T14:40:38.6701750Z SKIPPED 2025-09-09T14:40:38.6702214Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_09_cpu (m, k, n): (64, 256, 128) 2025-09-09T14:40:38.6702766Z SKIPPED 2025-09-09T14:40:38.6703228Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_10_cpu (m, k, n): (16, 128, 128) 2025-09-09T14:40:38.6703787Z SKIPPED 2025-09-09T14:40:38.6704261Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_11_cpu (m, k, n): (64, 128, 128) 2025-09-09T14:40:38.6704907Z SKIPPED 2025-09-09T14:40:38.6705377Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_12_cpu (m, k, n): (16, 128, 256) 2025-09-09T14:40:38.6705928Z SKIPPED 2025-09-09T14:40:38.6706476Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_13_cpu (m, k, n): (16, 256, 128) 2025-09-09T14:40:38.6707031Z SKIPPED 2025-09-09T14:40:38.6707499Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_14_cpu (m, k, n): (64, 256, 128) 2025-09-09T14:40:38.6708053Z SKIPPED 2025-09-09T14:40:38.6708523Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_15_cuda (m, k, n): (16, 128, 128) 2025-09-09T14:40:38.6709086Z PASSED 2025-09-09T14:40:38.6709552Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_16_cuda (m, k, n): (64, 128, 128) 2025-09-09T14:40:38.6710158Z activation_shapes: torch.Size([64, 128]), times_seen: 1 2025-09-09T14:40:38.6710634Z weight_shape: torch.Size([128, 128]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:40:38.6711073Z Autotune Choices Stats: 2025-09-09T14:40:38.6712028Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_981", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:40:38.6713031Z AUTOTUNE addmm(64x128, 64x128, 128x128) 2025-09-09T14:40:38.6713317Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:40:38.6713630Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:40:38.6714375Z triton_mm_981 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:40:38.6715391Z triton_mm_982 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6716488Z triton_mm_983 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:38.6717481Z triton_mm_984 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:38.6718467Z triton_mm_994 0.0276 ms 92.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:38.6719440Z triton_mm_987 0.0287 ms 89.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6720429Z triton_mm_986 0.0307 ms 83.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6721420Z triton_mm_990 0.0317 ms 80.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:40:38.6722400Z triton_mm_988 0.0358 ms 71.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:40:38.6723388Z triton_mm_995 0.0358 ms 71.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:40:38.6724256Z SingleProcess AUTOTUNE benchmarking takes 0.2748 seconds and 10.8658 seconds precompiling for 19 choices 2025-09-09T14:40:38.6724992Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:40:38.6725602Z Autotune Choices Stats: 2025-09-09T14:40:38.6726632Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1002", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:40:38.6727626Z AUTOTUNE mm(64x128, 128x128) 2025-09-09T14:40:38.6727877Z strides: [128, 1], [128, 1] 2025-09-09T14:40:38.6728135Z dtypes: torch.float32, torch.float32 2025-09-09T14:40:38.6728798Z triton_mm_1002 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:40:38.6729813Z triton_mm_1013 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:40:38.6730828Z triton_mm_1000 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:40:38.6731824Z triton_mm_1003 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6732812Z triton_mm_1004 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:40:38.6733809Z triton_mm_1007 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:01.6722370Z triton_mm_1008 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:01.6724905Z triton_mm_998 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:01.6725952Z triton_mm_999 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:01.6726967Z triton_mm_1001 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:01.6727854Z SingleProcess AUTOTUNE benchmarking takes 0.2371 seconds and 5.1183 seconds precompiling for 18 choices 2025-09-09T14:41:01.6728750Z >>time: 0.012ms for , to_beat: 0.014ms 2025-09-09T14:41:01.6729380Z Autotune Choices Stats: 2025-09-09T14:41:01.6730332Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_1019", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:41:01.6731333Z AUTOTUNE int_mm(64x128, 128x128) 2025-09-09T14:41:01.6731598Z strides: [128, 1], [1, 128] 2025-09-09T14:41:01.6731851Z dtypes: torch.int8, torch.int8 2025-09-09T14:41:01.6732481Z triton_mm_1019 0.0225 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:01.6733480Z triton_mm_1024 0.0236 ms 95.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:41:01.6734464Z triton_mm_1015 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:01.6735712Z triton_mm_1016 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:01.6736849Z triton_mm_1017 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:01.6737828Z triton_mm_1018 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:01.6738796Z triton_mm_1020 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:01.6739763Z triton_mm_1021 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:01.6740752Z triton_mm_1022 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:41:01.6741718Z triton_mm_1023 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:41:01.6742590Z SingleProcess AUTOTUNE benchmarking takes 0.1381 seconds and 0.3637 seconds precompiling for 11 choices 2025-09-09T14:41:01.6743463Z >>time: 0.008ms for matmul, to_beat: 0.012ms 2025-09-09T14:41:01.6744414Z >>time: 0.009ms for , to_beat: 0.035ms 2025-09-09T14:41:01.6745438Z >>time: 0.009ms for interpolated, breakeven constant: 3.55 2025-09-09T14:41:01.6746368Z best_cls= 2025-09-09T14:41:01.6746799Z 2025-09-09T14:41:01.6747098Z PASSED 2025-09-09T14:41:01.6747618Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_17_cuda (m, k, n): (16, 128, 256) 2025-09-09T14:41:01.6748225Z activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:41:01.6748711Z weight_shape: torch.Size([256, 128]), dtype: torch.float32, bias_shape: torch.Size([256]) 2025-09-09T14:41:01.6749141Z Autotune Choices Stats: 2025-09-09T14:41:01.6750110Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1036", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:41:01.6751120Z AUTOTUNE addmm(16x256, 16x128, 128x256) 2025-09-09T14:41:01.6751395Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:41:01.6751714Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:41:01.6752411Z triton_mm_1036 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:41:01.6753432Z triton_mm_1041 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:01.6754437Z triton_mm_1042 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:01.6755432Z triton_mm_1048 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:01.6756882Z triton_mm_1049 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:41:01.6758458Z triton_mm_1035 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:01.6759624Z triton_mm_1037 0.0256 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:01.6760628Z triton_mm_1040 0.0257 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:01.6761636Z triton_mm_1043 0.0257 ms 99.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:01.6762642Z triton_mm_1038 0.0266 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:01.6763528Z SingleProcess AUTOTUNE benchmarking takes 0.2370 seconds and 2.0010 seconds precompiling for 19 choices 2025-09-09T14:41:01.6764757Z >>time: 0.008ms for , to_beat: infms 2025-09-09T14:41:01.6765323Z Autotune Choices Stats: 2025-09-09T14:41:01.6766288Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1052", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:41:01.6767277Z AUTOTUNE mm(16x128, 128x256) 2025-09-09T14:41:01.6767533Z strides: [128, 1], [256, 1] 2025-09-09T14:41:01.6767796Z dtypes: torch.float32, torch.float32 2025-09-09T14:41:01.6768450Z triton_mm_1052 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:01.6769469Z triton_mm_1054 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:01.6770472Z triton_mm_1055 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:01.6771481Z triton_mm_1056 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:01.6772503Z triton_mm_1057 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:01.6773512Z triton_mm_1058 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:22.1438234Z triton_mm_1059 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1439295Z triton_mm_1060 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:22.1440316Z triton_mm_1061 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1441707Z triton_mm_1062 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:22.1443955Z SingleProcess AUTOTUNE benchmarking takes 0.9141 seconds and 0.8149 seconds precompiling for 18 choices 2025-09-09T14:41:22.1444838Z >>time: 0.012ms for , to_beat: 0.008ms 2025-09-09T14:41:22.1445756Z >>time: 0.014ms for , to_beat: 0.008ms 2025-09-09T14:41:22.1446360Z Autotune Choices Stats: 2025-09-09T14:41:22.1447311Z {"num_choices": 12, "num_triton_choices": 11, "best_kernel": "triton_mm_1070", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:41:22.1448308Z AUTOTUNE int_mm(16x128, 128x256) 2025-09-09T14:41:22.1448566Z strides: [128, 1], [1, 128] 2025-09-09T14:41:22.1448808Z dtypes: torch.int8, torch.int8 2025-09-09T14:41:22.1449438Z triton_mm_1070 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1450426Z triton_mm_1071 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1451402Z triton_mm_1072 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:22.1452432Z triton_mm_1074 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:22.1453412Z triton_mm_1075 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:22.1454381Z triton_mm_1077 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1455361Z triton_mm_1078 0.0246 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=256, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:41:22.1456348Z triton_mm_1073 0.0246 ms 99.9% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:41:22.1457306Z triton_mm_1069 0.0256 ms 96.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:22.1458288Z triton_mm_1076 0.0256 ms 96.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:41:22.1459149Z SingleProcess AUTOTUNE benchmarking takes 0.1417 seconds and 0.2953 seconds precompiling for 12 choices 2025-09-09T14:41:22.1460000Z >>time: 0.014ms for matmul, to_beat: 0.008ms 2025-09-09T14:41:22.1460776Z best_cls= 2025-09-09T14:41:22.1461107Z 2025-09-09T14:41:22.1461417Z PASSED 2025-09-09T14:41:22.1461972Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_18_cuda (m, k, n): (16, 256, 128) 2025-09-09T14:41:22.1462569Z activation_shapes: torch.Size([16, 256]), times_seen: 1 2025-09-09T14:41:22.1463037Z weight_shape: torch.Size([128, 256]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:41:22.1463564Z Autotune Choices Stats: 2025-09-09T14:41:22.1464834Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1081", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.025599999353289604, "best_triton_pos": 0} 2025-09-09T14:41:22.1465840Z AUTOTUNE addmm(16x128, 16x256, 256x128) 2025-09-09T14:41:22.1466111Z strides: [0, 1], [256, 1], [1, 256] 2025-09-09T14:41:22.1466420Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:41:22.1467112Z triton_mm_1081 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:41:22.1468118Z triton_mm_1083 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:22.1469126Z triton_mm_1084 0.0256 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:22.1470127Z triton_mm_1082 0.0266 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:22.1471109Z triton_mm_1093 0.0266 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1472147Z triton_mm_1087 0.0276 ms 92.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:22.1473130Z triton_mm_1094 0.0276 ms 92.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:41:22.1474116Z triton_mm_1080 0.0287 ms 89.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:22.1475108Z triton_mm_1086 0.0297 ms 86.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:22.1476092Z triton_mm_1085 0.0338 ms 75.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:22.1477007Z SingleProcess AUTOTUNE benchmarking takes 0.2168 seconds and 1.4272 seconds precompiling for 19 choices 2025-09-09T14:41:22.1477745Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:41:22.1478243Z Autotune Choices Stats: 2025-09-09T14:41:22.1479201Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1101", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:41:22.1480185Z AUTOTUNE mm(16x256, 256x128) 2025-09-09T14:41:22.1480424Z strides: [256, 1], [128, 1] 2025-09-09T14:41:22.1480681Z dtypes: torch.float32, torch.float32 2025-09-09T14:41:22.1481324Z triton_mm_1101 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:22.1482382Z triton_mm_1102 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:22.1483378Z triton_mm_1097 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:22.1484502Z triton_mm_1098 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:41:22.1485565Z triton_mm_1099 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:22.1486557Z triton_mm_1100 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:22.1487540Z triton_mm_1103 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:55.2683376Z triton_mm_1104 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2684452Z triton_mm_1106 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2685469Z triton_mm_1108 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2686341Z SingleProcess AUTOTUNE benchmarking takes 0.2000 seconds and 1.0925 seconds precompiling for 18 choices 2025-09-09T14:41:55.2687185Z >>time: 0.013ms for , to_beat: 0.011ms 2025-09-09T14:41:55.2688101Z >>time: 0.014ms for , to_beat: 0.011ms 2025-09-09T14:41:55.2688720Z Autotune Choices Stats: 2025-09-09T14:41:55.2689693Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_1123", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:41:55.2690690Z AUTOTUNE int_mm(16x256, 256x128) 2025-09-09T14:41:55.2690950Z strides: [256, 1], [1, 256] 2025-09-09T14:41:55.2691206Z dtypes: torch.int8, torch.int8 2025-09-09T14:41:55.2691827Z triton_mm_1123 0.0225 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:41:55.2692840Z triton_mm_1120 0.0236 ms 95.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:41:55.2693845Z triton_mm_1114 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:55.2694823Z triton_mm_1115 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2695791Z triton_mm_1116 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2696759Z triton_mm_1117 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:55.2697729Z triton_mm_1119 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:41:55.2699071Z triton_mm_1121 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:41:55.2700189Z triton_mm_1122 0.0246 ms 91.7% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2701159Z triton_mm_1118 0.0256 ms 88.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:41:55.2702015Z SingleProcess AUTOTUNE benchmarking takes 0.1313 seconds and 0.3022 seconds precompiling for 11 choices 2025-09-09T14:41:55.2702886Z >>time: 0.015ms for matmul, to_beat: 0.011ms 2025-09-09T14:41:55.2703695Z best_cls= 2025-09-09T14:41:55.2704037Z 2025-09-09T14:41:55.2704346Z PASSED 2025-09-09T14:41:55.2704845Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_19_cuda (m, k, n): (64, 256, 128) 2025-09-09T14:41:55.2705447Z activation_shapes: torch.Size([64, 256]), times_seen: 1 2025-09-09T14:41:55.2705925Z weight_shape: torch.Size([128, 256]), dtype: torch.float32, bias_shape: torch.Size([128]) 2025-09-09T14:41:55.2706365Z Autotune Choices Stats: 2025-09-09T14:41:55.2707321Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1125", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.027615999802947044, "best_triton_pos": 0} 2025-09-09T14:41:55.2708328Z AUTOTUNE addmm(64x128, 64x256, 256x128) 2025-09-09T14:41:55.2708600Z strides: [0, 1], [256, 1], [1, 256] 2025-09-09T14:41:55.2708902Z dtypes: torch.float32, torch.float32, torch.float32 2025-09-09T14:41:55.2709605Z triton_mm_1125 0.0276 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:55.2710611Z triton_mm_1124 0.0348 ms 79.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:55.2711606Z triton_mm_1126 0.0379 ms 72.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:55.2712602Z triton_mm_1127 0.0379 ms 72.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:55.2713638Z triton_mm_1137 0.0410 ms 67.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2714632Z triton_mm_1129 0.0451 ms 61.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:55.2715625Z triton_mm_1130 0.0451 ms 61.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:55.2716686Z triton_mm_1135 0.0451 ms 61.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2717685Z triton_mm_1133 0.0461 ms 59.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2718683Z triton_mm_1138 0.0471 ms 58.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:55.2719639Z SingleProcess AUTOTUNE benchmarking takes 0.2167 seconds and 4.5811 seconds precompiling for 19 choices 2025-09-09T14:41:55.2720375Z >>time: 0.017ms for , to_beat: infms 2025-09-09T14:41:55.2721498Z Autotune Choices Stats: 2025-09-09T14:41:55.2722470Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1154", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:41:55.2723493Z AUTOTUNE mm(64x256, 256x128) 2025-09-09T14:41:55.2723739Z strides: [256, 1], [128, 1] 2025-09-09T14:41:55.2723987Z dtypes: torch.float32, torch.float32 2025-09-09T14:41:55.2724637Z triton_mm_1154 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:41:55.2725658Z triton_mm_1155 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:41:55.2726658Z triton_mm_1148 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:41:55.2727657Z triton_mm_1142 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:41:55.2728661Z triton_mm_1144 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:55.2729648Z triton_mm_1143 0.0267 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:41:55.2730653Z triton_mm_1141 0.0276 ms 88.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:41:55.2731650Z triton_mm_1147 0.0276 ms 88.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:10.4681811Z triton_mm_1146 0.0317 ms 77.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:10.4682885Z triton_mm_1150 0.0317 ms 77.4% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:10.4683767Z SingleProcess AUTOTUNE benchmarking takes 0.2020 seconds and 3.7764 seconds precompiling for 18 choices 2025-09-09T14:42:10.4684632Z >>time: 0.018ms for , to_beat: 0.017ms 2025-09-09T14:42:10.4685233Z Autotune Choices Stats: 2025-09-09T14:42:10.4686187Z {"num_choices": 11, "num_triton_choices": 10, "best_kernel": "triton_mm_1160", "best_kernel_desc": "ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:42:10.4687243Z AUTOTUNE int_mm(64x256, 256x128) 2025-09-09T14:42:10.4687505Z strides: [256, 1], [1, 256] 2025-09-09T14:42:10.4687751Z dtypes: torch.int8, torch.int8 2025-09-09T14:42:10.4688376Z triton_mm_1160 0.0236 ms 100.0% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:10.4689353Z triton_mm_1158 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:10.4690843Z triton_mm_1159 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:10.4691824Z triton_mm_1161 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:42:10.4692847Z triton_mm_1162 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:42:10.4693809Z triton_mm_1163 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:10.4694786Z triton_mm_1165 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:42:10.4695761Z triton_mm_1166 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:42:10.4696734Z triton_mm_1167 0.0246 ms 95.8% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:42:10.4697706Z triton_mm_1164 0.0256 ms 92.1% ACC_TYPE='tl.int32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:10.4698558Z SingleProcess AUTOTUNE benchmarking takes 0.1426 seconds and 0.3667 seconds precompiling for 11 choices 2025-09-09T14:42:10.4699421Z >>time: 0.009ms for matmul, to_beat: 0.017ms 2025-09-09T14:42:10.4700388Z >>time: 0.011ms for , to_beat: 0.064ms 2025-09-09T14:42:10.4701407Z >>time: 0.011ms for interpolated, breakeven constant: 4.00 2025-09-09T14:42:10.4702333Z best_cls= 2025-09-09T14:42:10.4702808Z 2025-09-09T14:42:10.4703125Z PASSED 2025-09-09T14:42:10.4703629Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_20_cuda (m, k, n): (16, 128, 128) 2025-09-09T14:42:10.4704197Z PASSED 2025-09-09T14:42:10.4704667Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_21_cuda (m, k, n): (64, 128, 128) 2025-09-09T14:42:10.4705268Z activation_shapes: torch.Size([64, 128]), times_seen: 1 2025-09-09T14:42:10.4705749Z weight_shape: torch.Size([128, 128]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:42:10.4706185Z Autotune Choices Stats: 2025-09-09T14:42:10.4707158Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1186", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:42:10.4708159Z AUTOTUNE addmm(64x128, 64x128, 128x128) 2025-09-09T14:42:10.4708443Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:42:10.4708743Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:42:10.4709439Z triton_mm_1186 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:42:10.4710459Z triton_mm_1190 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:42:10.4711555Z triton_mm_1185 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:42:10.4712625Z triton_mm_1189 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:10.4713623Z triton_mm_1193 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:42:10.4714611Z triton_mm_1183 0.0257 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:10.4715604Z triton_mm_1192 0.0257 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:42:10.4716680Z triton_mm_1180 0.0266 ms 92.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:10.4717674Z triton_mm_1181 0.0266 ms 92.5% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:10.4718665Z triton_mm_1178 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:42:10.4719524Z SingleProcess AUTOTUNE benchmarking takes 0.2472 seconds and 0.5873 seconds precompiling for 19 choices 2025-09-09T14:42:10.4720261Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:42:10.4720772Z Autotune Choices Stats: 2025-09-09T14:42:10.4721726Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1211", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:42:10.4722761Z AUTOTUNE mm(64x128, 128x128) 2025-09-09T14:42:10.4723002Z strides: [128, 1], [128, 1] 2025-09-09T14:42:10.4723261Z dtypes: torch.float16, torch.float16 2025-09-09T14:42:10.4723913Z triton_mm_1211 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:10.4724920Z triton_mm_1203 0.0226 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:42:10.4725929Z triton_mm_1201 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:10.4726930Z triton_mm_1204 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:10.4727920Z triton_mm_1205 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:42:10.4728922Z triton_mm_1206 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:10.4729924Z triton_mm_1207 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:42:10.4731014Z triton_mm_1195 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:42:42.1854296Z triton_mm_1196 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:42.1855345Z triton_mm_1197 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:42.1856211Z SingleProcess AUTOTUNE benchmarking takes 0.2265 seconds and 0.4633 seconds precompiling for 18 choices 2025-09-09T14:42:42.1857044Z >>time: 0.010ms for , to_beat: 0.011ms 2025-09-09T14:42:42.1857980Z >>time: 0.008ms for matmul, to_beat: 0.010ms 2025-09-09T14:42:42.1858934Z >>time: 0.010ms for , to_beat: 0.022ms 2025-09-09T14:42:42.1859942Z >>time: 0.010ms for interpolated, breakeven constant: 1.00 2025-09-09T14:42:42.1860848Z best_cls= 2025-09-09T14:42:42.1861278Z 2025-09-09T14:42:42.1861578Z PASSED 2025-09-09T14:42:42.1862077Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_22_cuda (m, k, n): (16, 128, 256) 2025-09-09T14:42:42.1862674Z activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:42:42.1863174Z weight_shape: torch.Size([256, 128]), dtype: torch.float16, bias_shape: torch.Size([256]) 2025-09-09T14:42:42.1863615Z Autotune Choices Stats: 2025-09-09T14:42:42.1864813Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1245", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:42:42.1865818Z AUTOTUNE addmm(16x256, 16x128, 128x256) 2025-09-09T14:42:42.1866094Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:42:42.1866398Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:42:42.1867081Z triton_mm_1245 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:42.1868077Z triton_mm_1232 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:42:42.1869065Z triton_mm_1233 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:42:42.1870046Z triton_mm_1235 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:42:42.1871048Z triton_mm_1236 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:42:42.1872031Z triton_mm_1240 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:42:42.1873013Z triton_mm_1241 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:42.1874000Z triton_mm_1244 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:42:42.1877221Z triton_mm_1246 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:42:42.1878323Z triton_mm_1247 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:42:42.1879192Z SingleProcess AUTOTUNE benchmarking takes 0.2419 seconds and 0.3981 seconds precompiling for 19 choices 2025-09-09T14:42:42.1879919Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:42:42.1880412Z Autotune Choices Stats: 2025-09-09T14:42:42.1881354Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1251", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:42:42.1882325Z AUTOTUNE mm(16x128, 128x256) 2025-09-09T14:42:42.1882573Z strides: [128, 1], [256, 1] 2025-09-09T14:42:42.1882834Z dtypes: torch.float16, torch.float16 2025-09-09T14:42:42.1883471Z triton_mm_1251 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:42:42.1884464Z triton_mm_1252 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:42:42.1885447Z triton_mm_1262 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:42.1886451Z triton_mm_1264 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:42:42.1887450Z triton_mm_1265 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:42:42.1888429Z triton_mm_1249 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:42:42.1889411Z triton_mm_1253 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:42:42.1890388Z triton_mm_1254 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:42.1891360Z triton_mm_1255 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:42:42.1892341Z triton_mm_1256 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:42:42.1893245Z SingleProcess AUTOTUNE benchmarking takes 0.2231 seconds and 0.3318 seconds precompiling for 18 choices 2025-09-09T14:42:42.1894065Z >>time: 0.015ms for , to_beat: 0.014ms 2025-09-09T14:42:42.1894985Z >>time: 0.015ms for , to_beat: 0.014ms 2025-09-09T14:42:42.1895923Z >>time: 0.014ms for matmul, to_beat: 0.014ms 2025-09-09T14:42:42.1896957Z >>time: 0.009ms for , to_beat: 0.015ms 2025-09-09T14:42:42.1897975Z >>time: 0.010ms for interpolated, breakeven constant: -0.01 2025-09-09T14:42:42.1898976Z best_cls= 2025-09-09T14:42:42.1899408Z 2025-09-09T14:42:42.1899539Z PASSED 2025-09-09T14:42:42.1900024Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_23_cuda (m, k, n): (16, 256, 128) 2025-09-09T14:42:42.1900617Z activation_shapes: torch.Size([16, 256]), times_seen: 1 2025-09-09T14:42:42.1901090Z weight_shape: torch.Size([128, 256]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:42:42.1901508Z Autotune Choices Stats: 2025-09-09T14:42:42.1902472Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1289", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:42:42.1903524Z AUTOTUNE addmm(16x128, 16x256, 256x128) 2025-09-09T14:42:42.1903801Z strides: [0, 1], [256, 1], [1, 256] 2025-09-09T14:42:42.1904108Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:42:42.1904795Z triton_mm_1289 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:42:42.1905803Z triton_mm_1292 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:42:42.1906807Z triton_mm_1293 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8516479Z triton_mm_1294 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8518980Z triton_mm_1291 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:03.8520039Z triton_mm_1297 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8521048Z triton_mm_1298 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:43:03.8522047Z triton_mm_1299 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8523063Z triton_mm_1302 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:43:03.8524064Z triton_mm_1304 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:03.8524934Z SingleProcess AUTOTUNE benchmarking takes 0.2319 seconds and 0.4233 seconds precompiling for 19 choices 2025-09-09T14:43:03.8525680Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:43:03.8526187Z Autotune Choices Stats: 2025-09-09T14:43:03.8527205Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1309", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:43:03.8528512Z AUTOTUNE mm(16x256, 256x128) 2025-09-09T14:43:03.8528759Z strides: [256, 1], [128, 1] 2025-09-09T14:43:03.8529025Z dtypes: torch.float16, torch.float16 2025-09-09T14:43:03.8529853Z triton_mm_1309 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:03.8530872Z triton_mm_1306 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:43:03.8531864Z triton_mm_1305 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:03.8532844Z triton_mm_1307 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:03.8533841Z triton_mm_1308 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:03.8534837Z triton_mm_1310 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8535814Z triton_mm_1311 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8536818Z triton_mm_1312 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8537846Z triton_mm_1313 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:03.8538842Z triton_mm_1314 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8539730Z SingleProcess AUTOTUNE benchmarking takes 0.2120 seconds and 0.5239 seconds precompiling for 18 choices 2025-09-09T14:43:03.8540561Z >>time: 0.014ms for , to_beat: 0.014ms 2025-09-09T14:43:03.8548377Z >>time: 0.014ms for , to_beat: 0.014ms 2025-09-09T14:43:03.8549359Z >>time: 0.012ms for matmul, to_beat: 0.014ms 2025-09-09T14:43:03.8550323Z >>time: 0.012ms for , to_beat: 0.022ms 2025-09-09T14:43:03.8551357Z >>time: 0.012ms for interpolated, breakeven constant: -1.80 2025-09-09T14:43:03.8552297Z best_cls= 2025-09-09T14:43:03.8552726Z 2025-09-09T14:43:03.8553023Z PASSED 2025-09-09T14:43:03.8553531Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_24_cuda (m, k, n): (64, 256, 128) 2025-09-09T14:43:03.8554134Z activation_shapes: torch.Size([64, 256]), times_seen: 1 2025-09-09T14:43:03.8554610Z weight_shape: torch.Size([128, 256]), dtype: torch.float16, bias_shape: torch.Size([128]) 2025-09-09T14:43:03.8555043Z Autotune Choices Stats: 2025-09-09T14:43:03.8556014Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1351", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:43:03.8557271Z AUTOTUNE addmm(64x128, 64x256, 256x128) 2025-09-09T14:43:03.8557555Z strides: [0, 1], [256, 1], [1, 256] 2025-09-09T14:43:03.8557947Z dtypes: torch.float16, torch.float16, torch.float16 2025-09-09T14:43:03.8558656Z triton_mm_1351 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8559676Z triton_mm_1353 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8560674Z triton_mm_1345 0.0246 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:03.8561671Z triton_mm_1342 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:03.8562673Z triton_mm_1343 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8563672Z triton_mm_1344 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:03.8564863Z triton_mm_1346 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:03.8565858Z triton_mm_1347 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8566886Z triton_mm_1348 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:03.8567893Z triton_mm_1349 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:43:03.8568762Z SingleProcess AUTOTUNE benchmarking takes 0.2247 seconds and 0.6683 seconds precompiling for 19 choices 2025-09-09T14:43:03.8569499Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:43:03.8570004Z Autotune Choices Stats: 2025-09-09T14:43:03.8570960Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1370", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:43:03.8571954Z AUTOTUNE mm(64x256, 256x128) 2025-09-09T14:43:03.8572209Z strides: [256, 1], [128, 1] 2025-09-09T14:43:03.8572466Z dtypes: torch.float16, torch.float16 2025-09-09T14:43:03.8573129Z triton_mm_1370 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:03.8574140Z triton_mm_1362 0.0236 ms 99.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:16.9369934Z triton_mm_1359 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:16.9370970Z triton_mm_1361 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:16.9372348Z triton_mm_1363 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:16.9373520Z triton_mm_1365 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:16.9374513Z triton_mm_1366 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:43:16.9375511Z triton_mm_1368 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9376549Z triton_mm_1369 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:43:16.9377551Z triton_mm_1372 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9378420Z SingleProcess AUTOTUNE benchmarking takes 0.2053 seconds and 0.5691 seconds precompiling for 18 choices 2025-09-09T14:43:16.9379259Z >>time: 0.012ms for , to_beat: 0.011ms 2025-09-09T14:43:16.9380210Z >>time: 0.009ms for matmul, to_beat: 0.011ms 2025-09-09T14:43:16.9381153Z >>time: 0.011ms for , to_beat: 0.023ms 2025-09-09T14:43:16.9382168Z >>time: 0.011ms for interpolated, breakeven constant: 1.00 2025-09-09T14:43:16.9383087Z best_cls= 2025-09-09T14:43:16.9383519Z 2025-09-09T14:43:16.9383830Z PASSED 2025-09-09T14:43:16.9384338Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_25_cuda (m, k, n): (16, 128, 128) 2025-09-09T14:43:16.9384900Z PASSED 2025-09-09T14:43:16.9385373Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_26_cuda (m, k, n): (64, 128, 128) 2025-09-09T14:43:16.9385963Z activation_shapes: torch.Size([64, 128]), times_seen: 1 2025-09-09T14:43:16.9386447Z weight_shape: torch.Size([128, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:43:16.9386876Z Autotune Choices Stats: 2025-09-09T14:43:16.9387837Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1405", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02457600086927414, "best_triton_pos": 0} 2025-09-09T14:43:16.9388856Z AUTOTUNE addmm(64x128, 64x128, 128x128) 2025-09-09T14:43:16.9389126Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:43:16.9389454Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:43:16.9390180Z triton_mm_1405 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9391199Z triton_mm_1408 0.0246 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:43:16.9392205Z triton_mm_1407 0.0255 ms 96.2% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9395303Z triton_mm_1396 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:16.9396488Z triton_mm_1398 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:16.9397488Z triton_mm_1404 0.0256 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:16.9398481Z triton_mm_1406 0.0256 ms 95.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:43:16.9399467Z triton_mm_1401 0.0257 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:16.9400459Z triton_mm_1409 0.0257 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9401454Z triton_mm_1397 0.0266 ms 92.3% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:16.9402317Z SingleProcess AUTOTUNE benchmarking takes 0.2410 seconds and 0.5726 seconds precompiling for 19 choices 2025-09-09T14:43:16.9403055Z >>time: 0.009ms for , to_beat: infms 2025-09-09T14:43:16.9403557Z Autotune Choices Stats: 2025-09-09T14:43:16.9404826Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1422", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:43:16.9405874Z AUTOTUNE mm(64x128, 128x128) 2025-09-09T14:43:16.9406117Z strides: [128, 1], [128, 1] 2025-09-09T14:43:16.9406373Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:43:16.9407033Z triton_mm_1422 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9408040Z triton_mm_1424 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:16.9409049Z triton_mm_1425 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:43:16.9410046Z triton_mm_1417 0.0245 ms 96.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:16.9411044Z triton_mm_1413 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:16.9412037Z triton_mm_1414 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:16.9413017Z triton_mm_1416 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:43:16.9414003Z triton_mm_1418 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:16.9414984Z triton_mm_1419 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:16.9416304Z triton_mm_1420 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:43:16.9417436Z SingleProcess AUTOTUNE benchmarking takes 0.2221 seconds and 0.4965 seconds precompiling for 18 choices 2025-09-09T14:43:16.9418441Z >>time: 0.009ms for , to_beat: 0.009ms 2025-09-09T14:43:16.9419603Z >>time: 0.014ms for matmul, to_beat: 0.009ms 2025-09-09T14:43:16.9420533Z best_cls= 2025-09-09T14:43:16.9420926Z 2025-09-09T14:43:16.9421066Z PASSED 2025-09-09T14:43:16.9421550Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_27_cuda (m, k, n): (16, 128, 256) 2025-09-09T14:43:16.9422146Z activation_shapes: torch.Size([16, 128]), times_seen: 1 2025-09-09T14:43:16.9422625Z weight_shape: torch.Size([256, 128]), dtype: torch.bfloat16, bias_shape: torch.Size([256]) 2025-09-09T14:43:16.9423056Z Autotune Choices Stats: 2025-09-09T14:43:42.2298675Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1453", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:43:42.2300740Z AUTOTUNE addmm(16x256, 16x128, 128x256) 2025-09-09T14:43:42.2301231Z strides: [0, 1], [128, 1], [1, 128] 2025-09-09T14:43:42.2301599Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:43:42.2302310Z triton_mm_1453 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:42.2303326Z triton_mm_1454 0.0236 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:43:42.2304315Z triton_mm_1447 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:42.2305301Z triton_mm_1448 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:42.2306290Z triton_mm_1449 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:42.2307269Z triton_mm_1450 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:43:42.2308261Z triton_mm_1451 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:42.2309261Z triton_mm_1452 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:43:42.2310249Z triton_mm_1455 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:43:42.2311242Z triton_mm_1443 0.0246 ms 95.6% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:42.2312147Z SingleProcess AUTOTUNE benchmarking takes 0.2432 seconds and 0.4302 seconds precompiling for 19 choices 2025-09-09T14:43:42.2313146Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:43:42.2313648Z Autotune Choices Stats: 2025-09-09T14:43:42.2314745Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1460", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:43:42.2315717Z AUTOTUNE mm(16x128, 128x256) 2025-09-09T14:43:42.2315963Z strides: [128, 1], [256, 1] 2025-09-09T14:43:42.2316221Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:43:42.2316956Z triton_mm_1460 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:42.2317944Z triton_mm_1465 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:42.2318938Z triton_mm_1472 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:43:42.2319926Z triton_mm_1457 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:42.2320906Z triton_mm_1458 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:43:42.2321938Z triton_mm_1459 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:42.2322925Z triton_mm_1461 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:42.2323909Z triton_mm_1462 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:42.2324890Z triton_mm_1463 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:42.2325867Z triton_mm_1464 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:42.2326718Z SingleProcess AUTOTUNE benchmarking takes 0.2224 seconds and 0.4033 seconds precompiling for 18 choices 2025-09-09T14:43:42.2327548Z >>time: 0.014ms for , to_beat: 0.014ms 2025-09-09T14:43:42.2328468Z >>time: 0.016ms for , to_beat: 0.014ms 2025-09-09T14:43:42.2329423Z >>time: 0.008ms for matmul, to_beat: 0.014ms 2025-09-09T14:43:42.2330366Z >>time: 0.009ms for , to_beat: 0.052ms 2025-09-09T14:43:42.2331396Z >>time: 0.009ms for interpolated, breakeven constant: 4.33 2025-09-09T14:43:42.2332342Z best_cls= 2025-09-09T14:43:42.2332767Z 2025-09-09T14:43:42.2333090Z PASSED 2025-09-09T14:43:42.2333593Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_28_cuda (m, k, n): (16, 256, 128) 2025-09-09T14:43:42.2334289Z activation_shapes: torch.Size([16, 256]), times_seen: 1 2025-09-09T14:43:42.2334769Z weight_shape: torch.Size([128, 256]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:43:42.2335205Z Autotune Choices Stats: 2025-09-09T14:43:42.2336237Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1500", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:43:42.2337232Z AUTOTUNE addmm(16x128, 16x256, 256x128) 2025-09-09T14:43:42.2337515Z strides: [0, 1], [256, 1], [1, 256] 2025-09-09T14:43:42.2337833Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:43:42.2338538Z triton_mm_1500 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:43:42.2339548Z triton_mm_1504 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:42.2340552Z triton_mm_1498 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:43:42.2341572Z triton_mm_1501 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:43:42.2342569Z triton_mm_1505 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:43:42.2343565Z triton_mm_1506 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:43:42.2344562Z triton_mm_1510 0.0246 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:43:42.2345546Z triton_mm_1496 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:43:42.2346540Z triton_mm_1497 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=2 2025-09-09T14:43:42.2347527Z triton_mm_1499 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:44:06.0029350Z SingleProcess AUTOTUNE benchmarking takes 0.2314 seconds and 0.5207 seconds precompiling for 19 choices 2025-09-09T14:44:06.0030156Z >>time: 0.014ms for , to_beat: infms 2025-09-09T14:44:06.0030683Z Autotune Choices Stats: 2025-09-09T14:44:06.0031832Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1518", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:44:06.0032855Z AUTOTUNE mm(16x256, 256x128) 2025-09-09T14:44:06.0033103Z strides: [256, 1], [128, 1] 2025-09-09T14:44:06.0033361Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:44:06.0034033Z triton_mm_1518 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:44:06.0035044Z triton_mm_1519 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:44:06.0036622Z triton_mm_1521 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:44:06.0038888Z triton_mm_1523 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:44:06.0039909Z triton_mm_1524 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0040905Z triton_mm_1526 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0041905Z triton_mm_1527 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=4 2025-09-09T14:44:06.0042911Z triton_mm_1528 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:44:06.0043913Z triton_mm_1529 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=16, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:44:06.0044910Z triton_mm_1516 0.0236 ms 99.9% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=16, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=2 2025-09-09T14:44:06.0045775Z SingleProcess AUTOTUNE benchmarking takes 0.2084 seconds and 0.4231 seconds precompiling for 18 choices 2025-09-09T14:44:06.0046600Z >>time: 0.013ms for , to_beat: 0.014ms 2025-09-09T14:44:06.0047527Z >>time: 0.014ms for , to_beat: 0.013ms 2025-09-09T14:44:06.0048474Z >>time: 0.015ms for matmul, to_beat: 0.013ms 2025-09-09T14:44:06.0049321Z best_cls= 2025-09-09T14:44:06.0049737Z 2025-09-09T14:44:06.0050046Z PASSED 2025-09-09T14:44:06.0050541Z test/integration/test_integration.py::TestAutoQuant::test_autoquant_one_input_29_cuda (m, k, n): (64, 256, 128) 2025-09-09T14:44:06.0051139Z activation_shapes: torch.Size([64, 256]), times_seen: 1 2025-09-09T14:44:06.0051620Z weight_shape: torch.Size([128, 256]), dtype: torch.bfloat16, bias_shape: torch.Size([128]) 2025-09-09T14:44:06.0052052Z Autotune Choices Stats: 2025-09-09T14:44:06.0053007Z {"num_choices": 19, "num_triton_choices": 17, "best_kernel": "triton_mm_1549", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.023552000522613525, "best_triton_pos": 0} 2025-09-09T14:44:06.0053997Z AUTOTUNE addmm(64x128, 64x256, 256x128) 2025-09-09T14:44:06.0054277Z strides: [0, 1], [256, 1], [1, 256] 2025-09-09T14:44:06.0054595Z dtypes: torch.bfloat16, torch.bfloat16, torch.bfloat16 2025-09-09T14:44:06.0055296Z triton_mm_1549 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0056305Z triton_mm_1550 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=4, num_warps=8 2025-09-09T14:44:06.0057301Z triton_mm_1551 0.0236 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0058394Z triton_mm_1542 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:44:06.0059455Z triton_mm_1543 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:44:06.0060434Z triton_mm_1547 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:44:06.0061420Z triton_mm_1553 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0062404Z triton_mm_1555 0.0246 ms 95.8% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=8 2025-09-09T14:44:06.0063393Z triton_mm_1544 0.0256 ms 92.1% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:44:06.0064735Z triton_mm_1540 0.0256 ms 92.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=1, num_warps=2 2025-09-09T14:44:06.0065599Z SingleProcess AUTOTUNE benchmarking takes 0.2230 seconds and 0.6329 seconds precompiling for 19 choices 2025-09-09T14:44:06.0066323Z >>time: 0.011ms for , to_beat: infms 2025-09-09T14:44:06.0066824Z Autotune Choices Stats: 2025-09-09T14:44:06.0067762Z {"num_choices": 18, "num_triton_choices": 17, "best_kernel": "triton_mm_1570", "best_kernel_desc": "ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4", "best_time": 0.02252800017595291, "best_triton_pos": 0} 2025-09-09T14:44:06.0068735Z AUTOTUNE mm(64x256, 256x128) 2025-09-09T14:44:06.0068984Z strides: [256, 1], [128, 1] 2025-09-09T14:44:06.0069239Z dtypes: torch.bfloat16, torch.bfloat16 2025-09-09T14:44:06.0069896Z triton_mm_1570 0.0225 ms 100.0% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0070887Z triton_mm_1568 0.0236 ms 95.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0071879Z triton_mm_1558 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=32, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:44:06.0072863Z triton_mm_1559 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=32, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:44:06.0073845Z triton_mm_1560 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=8 2025-09-09T14:44:06.0074835Z triton_mm_1561 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=128, BLOCK_M=64, BLOCK_N=32, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=5, num_warps=4 2025-09-09T14:44:06.0075822Z triton_mm_1562 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=16, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:44:06.0076880Z triton_mm_1563 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=2, num_warps=4 2025-09-09T14:44:06.0078011Z triton_mm_1564 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=64, BLOCK_M=64, BLOCK_N=64, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=8 2025-09-09T14:44:06.0079107Z triton_mm_1566 0.0246 ms 91.7% ACC_TYPE='tl.float32', ALLOW_TF32=False, BLOCK_K=32, BLOCK_M=64, BLOCK_N=128, EVEN_K=True, GROUP_M=8, USE_FAST_ACCUM=False, num_stages=3, num_warps=4 2025-09-09T14:44:06.0079966Z SingleProcess AUTOTUNE benchmarking takes 0.2041 seconds and 0.5738 seconds precompiling for 18 choices 2025-09-09T14:44:20.0756604Z >>time: 0.010ms for , to_beat: 0.011ms 2025-09-09T14:44:20.0757852Z >>time: 0.009ms for matmul, to_beat: 0.010ms 2025-09-09T14:44:20.0759026Z >>time: 0.010ms for , to_beat: 0.019ms 2025-09-09T14:44:20.0760285Z >>time: 0.010ms for interpolated, breakeven constant: 0.97 2025-09-09T14:44:20.0761447Z best_cls= 2025-09-09T14:44:20.0761964Z 2025-09-09T14:44:20.0762317Z PASSED 2025-09-09T14:44:20.0762886Z test/integration/test_integration.py::TestAOTI::test_aoti_00 SKIPPED 2025-09-09T14:44:20.0763672Z test/integration/test_integration.py::TestAOTI::test_aoti_01 SKIPPED 2025-09-09T14:44:20.0764816Z test/integration/test_integration.py::TestAOTI::test_aoti_02 SKIPPED 2025-09-09T14:44:20.0765622Z test/integration/test_integration.py::TestAOTI::test_aoti_03 SKIPPED 2025-09-09T14:44:20.0766428Z test/integration/test_integration.py::TestAOTI::test_aoti_04 SKIPPED 2025-09-09T14:44:20.0767203Z test/integration/test_integration.py::TestAOTI::test_aoti_05 SKIPPED 2025-09-09T14:44:20.0767988Z test/integration/test_integration.py::TestAOTI::test_aoti_06 SKIPPED 2025-09-09T14:44:20.0768777Z test/integration/test_integration.py::TestAOTI::test_aoti_07 SKIPPED 2025-09-09T14:44:20.0769569Z test/integration/test_integration.py::TestAOTI::test_aoti_08 SKIPPED 2025-09-09T14:44:20.0770355Z test/integration/test_integration.py::TestAOTI::test_aoti_09 SKIPPED 2025-09-09T14:44:20.0771140Z test/integration/test_integration.py::TestAOTI::test_aoti_10 SKIPPED 2025-09-09T14:44:20.0771924Z test/integration/test_integration.py::TestAOTI::test_aoti_11 SKIPPED 2025-09-09T14:44:20.0772712Z test/integration/test_integration.py::TestAOTI::test_aoti_12 SKIPPED 2025-09-09T14:44:20.0773497Z test/integration/test_integration.py::TestAOTI::test_aoti_13 SKIPPED 2025-09-09T14:44:20.0774280Z test/integration/test_integration.py::TestAOTI::test_aoti_14 SKIPPED 2025-09-09T14:44:20.0775069Z test/integration/test_integration.py::TestAOTI::test_aoti_15 SKIPPED 2025-09-09T14:44:20.0775916Z test/integration/test_integration.py::TestAOTI::test_aoti_16 SKIPPED 2025-09-09T14:44:20.0776695Z test/integration/test_integration.py::TestAOTI::test_aoti_17 SKIPPED 2025-09-09T14:44:20.0777513Z test/integration/test_integration.py::TestExport::test_export_00 PASSED 2025-09-09T14:44:20.0778329Z test/integration/test_integration.py::TestExport::test_export_01 PASSED 2025-09-09T14:44:20.0779146Z test/integration/test_integration.py::TestExport::test_export_02 PASSED 2025-09-09T14:44:20.0779967Z test/integration/test_integration.py::TestExport::test_export_03 PASSED 2025-09-09T14:44:20.0780773Z test/integration/test_integration.py::TestExport::test_export_04 PASSED 2025-09-09T14:44:20.0781585Z test/integration/test_integration.py::TestExport::test_export_05 PASSED 2025-09-09T14:44:20.0782390Z test/integration/test_integration.py::TestExport::test_export_06 PASSED 2025-09-09T14:44:20.0783200Z test/integration/test_integration.py::TestExport::test_export_07 PASSED 2025-09-09T14:44:20.0784416Z test/integration/test_integration.py::TestExport::test_export_08 PASSED 2025-09-09T14:44:20.0785218Z test/integration/test_integration.py::TestExport::test_export_09 PASSED 2025-09-09T14:44:20.0786202Z test/integration/test_integration.py::TestExport::test_export_10 PASSED 2025-09-09T14:44:20.0786973Z test/integration/test_integration.py::TestExport::test_export_11 PASSED 2025-09-09T14:44:20.0787619Z test/integration/test_integration.py::TestExport::test_export_12 PASSED 2025-09-09T14:44:20.0788262Z test/integration/test_integration.py::TestExport::test_export_13 PASSED 2025-09-09T14:44:20.0788908Z test/integration/test_integration.py::TestExport::test_export_14 PASSED 2025-09-09T14:44:20.0789546Z test/integration/test_integration.py::TestExport::test_export_15 PASSED 2025-09-09T14:44:20.0790195Z test/integration/test_integration.py::TestExport::test_export_16 PASSED 2025-09-09T14:44:20.0790852Z test/integration/test_integration.py::TestExport::test_export_17 PASSED 2025-09-09T14:44:20.0791489Z test/integration/test_integration.py::TestExport::test_export_18 PASSED 2025-09-09T14:44:20.0792134Z test/integration/test_integration.py::TestExport::test_export_19 PASSED 2025-09-09T14:44:20.0792785Z test/integration/test_integration.py::TestExport::test_export_20 PASSED 2025-09-09T14:44:20.0793429Z test/integration/test_integration.py::TestExport::test_export_21 PASSED 2025-09-09T14:44:20.0794072Z test/integration/test_integration.py::TestExport::test_export_22 PASSED 2025-09-09T14:44:20.0794715Z test/integration/test_integration.py::TestExport::test_export_23 PASSED 2025-09-09T14:44:20.0795387Z test/integration/test_integration.py::TestExport::test_export_float8 SKIPPED 2025-09-09T14:44:20.0796100Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_00 SKIPPED 2025-09-09T14:44:20.0796903Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_01 SKIPPED 2025-09-09T14:44:20.0797629Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_02 SKIPPED 2025-09-09T14:44:20.0798361Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_03 SKIPPED 2025-09-09T14:44:20.0799094Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_04 SKIPPED 2025-09-09T14:44:20.0799813Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_05 PASSED 2025-09-09T14:44:20.0800540Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_06 SKIPPED 2025-09-09T14:44:20.0801263Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_07 SKIPPED 2025-09-09T14:44:20.0801993Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_08 SKIPPED 2025-09-09T14:44:20.0802721Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_09 SKIPPED 2025-09-09T14:44:20.0803445Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_10 SKIPPED 2025-09-09T14:44:20.0804179Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_11 PASSED 2025-09-09T14:44:20.0804908Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_12 SKIPPED 2025-09-09T14:44:20.0805634Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_13 SKIPPED 2025-09-09T14:44:20.0806359Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_14 SKIPPED 2025-09-09T14:44:20.0807081Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_15 SKIPPED 2025-09-09T14:44:20.0807815Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_16 SKIPPED 2025-09-09T14:44:20.0808535Z test/integration/test_integration.py::TestUtils::test_get_model_size_aqt_17 PASSED 2025-09-09T14:44:20.0809472Z test/integration/test_integration.py::TestBenchmarkModel::test_benchmark_model_cpu cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:44:20.0810270Z File "/pytorch/ao/test/integration/test_integration.py", line 2042, in forward 2025-09-09T14:44:20.0810691Z x = self.linear1(x) 2025-09-09T14:44:20.0812438Z 2025-09-09T14:44:20.0812605Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T14:44:20.0813077Z File "/pytorch/ao/test/integration/test_integration.py", line 2043, in forward 2025-09-09T14:44:20.0813497Z x = self.linear2(x) 2025-09-09T14:44:20.0813639Z 2025-09-09T14:44:20.0813762Z PASSED 2025-09-09T14:44:20.0814298Z test/integration/test_integration.py::TestBenchmarkModel::test_benchmark_model_cuda PASSED 2025-09-09T14:44:20.0815258Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_deprecated_hf_models_model_info0 SKIPPED 2025-09-09T14:44:20.0816762Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_deprecated_single_linear_model_name_torchao-testing/single-linear-Float8DynamicActivationFloat8WeightConfig-v1-0_13_dev SKIPPED 2025-09-09T14:44:20.0818611Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_single_linear_model_name_torchao-testing/single-linear-Float8DynamicActivationFloat8WeightConfig-v2-0_13_dev SKIPPED 2025-09-09T14:44:20.0820349Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_single_linear_model_name_torchao-testing/single-linear-Int4WeightOnlyConfig-preshuffled-v2-0_13_dev SKIPPED 2025-09-09T14:44:20.0821944Z test/integration/test_load_and_run_checkpoint.py::TestLoadAndRunCheckpoint::test_single_linear_model_name_torchao-testing/single-linear-Int4WeightOnlyConfig-v2-0_13_dev SKIPPED 2025-09-09T14:44:20.0823004Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_0_cuda PASSED 2025-09-09T14:44:20.0823643Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_1_cuda PASSED 2025-09-09T14:44:20.0824310Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_float8_0_cuda SKIPPED 2025-09-09T14:44:20.0825005Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_float8_1_cuda SKIPPED 2025-09-09T14:44:20.0825749Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_0_cuda PASSED 2025-09-09T14:44:20.0826430Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_1_cpu PASSED 2025-09-09T14:44:20.0827105Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_2_cuda PASSED 2025-09-09T14:44:20.0827790Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_3_cpu PASSED 2025-09-09T14:44:20.0828696Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-2-512-128] SKIPPED 2025-09-09T14:44:20.0829753Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-2-5120-1280] SKIPPED 2025-09-09T14:44:20.5261231Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-3-2048-2048] SKIPPED 2025-09-09T14:44:20.5262588Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-4-3584-640] SKIPPED 2025-09-09T14:44:20.5264113Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-13-8704-8576] SKIPPED 2025-09-09T14:44:20.5265485Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-26-18944-1664] SKIPPED 2025-09-09T14:44:20.5266998Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x128[dtype0-67-6656-1408] SKIPPED 2025-09-09T14:44:20.5268355Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-2-512-128] SKIPPED 2025-09-09T14:44:20.5269878Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-2-5120-1280] SKIPPED 2025-09-09T14:44:20.5271329Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-3-2048-2048] SKIPPED 2025-09-09T14:44:20.5272673Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-4-3584-640] SKIPPED 2025-09-09T14:44:20.5274007Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-13-8704-8576] SKIPPED 2025-09-09T14:44:20.5275364Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-26-18944-1664] SKIPPED 2025-09-09T14:44:20.5276777Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_fp8_gemm_1x128_128x1[dtype0-67-6656-1408] SKIPPED 2025-09-09T14:44:20.5278105Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_lhs[128] SKIPPED 2025-09-09T14:44:20.5279397Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_lhs[256] SKIPPED 2025-09-09T14:44:20.5280672Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_rhs[128] SKIPPED 2025-09-09T14:44:20.5281953Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_rhs[256] SKIPPED 2025-09-09T14:44:20.5283346Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-1024-128] SKIPPED 2025-09-09T14:44:20.5284800Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-1024-256] SKIPPED 2025-09-09T14:44:20.5286278Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-16384-128] SKIPPED 2025-09-09T14:44:20.5287755Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_act_quant_transposed_lhs[4096-16384-256] SKIPPED 2025-09-09T14:44:20.5289190Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-1024-128] SKIPPED 2025-09-09T14:44:20.5290592Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-1024-256] SKIPPED 2025-09-09T14:44:20.5291972Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-16384-128] SKIPPED 2025-09-09T14:44:20.5293361Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_rhs[4096-16384-256] SKIPPED 2025-09-09T14:44:20.5294767Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_transposed_rhs[128] SKIPPED 2025-09-09T14:44:20.5296164Z test/prototype/blockwise_fp8_training/test_blockwise_kernels.py::test_triton_quantize_fp8_weight_quant_transposed_rhs[256] SKIPPED 2025-09-09T14:44:20.5297529Z test/prototype/inductor/test_int8_sdpa_fusion.py::SDPAPatternRewriterCpuTests::test_sdpa_int8_rewriter_cpu SKIPPED 2025-09-09T14:44:20.5298830Z test/prototype/module_swap_quantization/test_kmeans_codebook.py::TestKmeansCodebook::test_kmeans_codebook SKIPPED 2025-09-09T14:44:20.5300076Z test/prototype/module_swap_quantization/test_llm_ptq_data_getter.py::TestPTQDataGetter::test_data_getter SKIPPED 2025-09-09T14:44:20.5301287Z test/prototype/module_swap_quantization/test_module_swap.py::TestEmbeddingSwap::test_embedding_swap PASSED 2025-09-09T14:44:20.5302668Z test/prototype/module_swap_quantization/test_module_swap_quantization_utils.py::TestQuantizedModuleUtils::test_set_bit_widths_by_name PASSED 2025-09-09T14:44:20.5304193Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantize_dynamic PASSED 2025-09-09T14:44:20.5305660Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantize_dynamic_vectorized PASSED 2025-09-09T14:44:20.5307045Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear PASSED 2025-09-09T14:44:20.5308356Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_init PASSED 2025-09-09T14:44:20.5309745Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_passes_gradients PASSED 2025-09-09T14:44:20.5311316Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_passes_gradients_to_activation_scale PASSED 2025-09-09T14:44:20.5312969Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_quantized_linear_passes_gradients_to_weight_scale PASSED 2025-09-09T14:44:20.5314581Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_set_weight_scale_to_min_max_test_all_options PASSED 2025-09-09T14:44:20.5316152Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedLinear::test_set_weight_scale_to_min_max_test_correct PASSED 2025-09-09T14:44:20.5317682Z test/prototype/module_swap_quantization/test_quantized_modules.py::TestQuantizedEmbedding::test_quantized_embedding PASSED 2025-09-09T14:44:20.5318925Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_qmin_qmax PASSED 2025-09-09T14:44:20.5320111Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_from_min_max PASSED 2025-09-09T14:44:20.5329025Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_from_min_max_vectorized PASSED 2025-09-09T14:44:20.5330089Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_asymmetric PASSED 2025-09-09T14:44:20.5331146Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_from_min_max PASSED 2025-09-09T14:44:20.5332235Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_from_min_max_tensorized PASSED 2025-09-09T14:44:20.5333300Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_offset_symmetric PASSED 2025-09-09T14:44:20.5334301Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_get_scale_param_size PASSED 2025-09-09T14:44:20.5335264Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward PASSED 2025-09-09T14:44:20.5336294Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward_asymmetric_clipping PASSED 2025-09-09T14:44:20.5337363Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward_symmetric PASSED 2025-09-09T14:44:20.5338425Z test/prototype/module_swap_quantization/test_quantizers.py::TestIntQuantizer::test_quantize_forward_symmetric_clipping PASSED 2025-09-09T14:44:20.5339483Z test/prototype/module_swap_quantization/test_quantizers.py::TestCodebookQuantizer::test_codebook_quantizer PASSED 2025-09-09T14:44:20.5340493Z test/prototype/module_swap_quantization/test_quantizers.py::TestCodebookQuantizer::test_vector_quantizer PASSED 2025-09-09T14:44:20.5341528Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMinMax::test_set_weight_min_max PASSED 2025-09-09T14:44:20.5342763Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMinMax::test_set_weight_min_max_grouped PASSED 2025-09-09T14:44:20.5343827Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMSE::test_set_weight_mse PASSED 2025-09-09T14:44:20.5344951Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightMSE::test_set_weight_mse_grouped PASSED 2025-09-09T14:44:20.5346212Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightRangeActivationLoss::test_set_weight_range_activation_loss PASSED 2025-09-09T14:44:20.6131278Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestSetWeightRangeActivationLoss::test_set_weight_range_activation_loss_progressive PASSED 2025-09-09T14:44:20.6132665Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestStaticActivationRangeSetting::test_static_activation_range_setting PASSED 2025-09-09T14:44:20.6134021Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestStaticActivationRangeSetting::test_static_activation_range_setting_no_input PASSED 2025-09-09T14:44:20.6135309Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestQuantizePerGroupScales::test_quantize_per_group_scales PASSED 2025-09-09T14:44:20.6136596Z test/prototype/module_swap_quantization/test_range_setting_methods.py::TestQuantizePerGroupScales::test_quantize_per_group_scales_dont_change_per_channel PASSED 2025-09-09T14:44:20.6137752Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[True-True-elem_dtype0] SKIPPED 2025-09-09T14:44:20.6138713Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[True-True-elem_dtype1] SKIPPED 2025-09-09T14:44:20.6139667Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[True-False-elem_dtype0] SKIPPED 2025-09-09T14:44:20.6140646Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[True-False-elem_dtype1] SKIPPED 2025-09-09T14:44:20.6141604Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[False-True-elem_dtype0] SKIPPED 2025-09-09T14:44:20.6142562Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[False-True-elem_dtype1] SKIPPED 2025-09-09T14:44:20.6143529Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[False-False-elem_dtype0] SKIPPED 2025-09-09T14:44:20.6144493Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_mx[False-False-elem_dtype1] SKIPPED 2025-09-09T14:44:20.6145654Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.6147037Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.6148354Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.6149679Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.6151015Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.6152352Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.6153683Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.6155213Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.6156758Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.6158084Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.6159400Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.6160728Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.6162058Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.6163384Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.6164882Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.6166264Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.6167593Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.6168933Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.6170269Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.6171596Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.6172941Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.6174289Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.6175631Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.6176981Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.6178318Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.6179648Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.6180976Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.6182304Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.6183776Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.6185231Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.6186573Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.6187920Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.6189263Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.6190592Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.6191932Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.6193273Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.6194613Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.6195998Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.6994586Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.6996022Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.6997474Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.6999228Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7000735Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7002320Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7003859Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.7005230Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7006572Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.7008126Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7009831Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7011683Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7013317Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7014678Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7016036Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.7017399Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7018760Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.7020125Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7021474Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7022803Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7024144Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7025503Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7026892Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.7028239Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7029598Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.7031113Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[256x128x512-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7032455Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7033795Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7035122Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7036560Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7037893Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.7039233Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7040683Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.7042100Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7043732Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7045354Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7047027Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7048663Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7050292Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.7051621Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7052963Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.7054309Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7055644Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7057046Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7058380Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7059722Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7061070Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.7062418Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7064000Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.7821448Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7822798Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7824140Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7825482Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7827015Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7828474Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.7829823Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7831167Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.7832518Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[145x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7833859Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7835185Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7836585Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7837914Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7839244Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.7840592Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7841929Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.7843263Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7844588Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7845898Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7847223Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7848552Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7849871Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.7851200Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7852531Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.7853852Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7855316Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7856721Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7858052Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7859390Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7860721Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.7862079Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7863435Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.7864933Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7866321Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7867645Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7868970Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7870316Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7871652Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.7872981Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7874324Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.7875668Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x96x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.7877119Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.7878450Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.7879781Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.7881113Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.7882451Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.7883938Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.7885394Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.7886747Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8647593Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.8648968Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.8650330Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.8651754Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.8653250Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.8654599Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.8656184Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.8657529Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8658888Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.8660234Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.8661784Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.8663148Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.8664859Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.8666245Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.8667617Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.8668974Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8670323Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.8671671Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.8673200Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.8674670Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.8676015Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.8677434Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.8678795Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.8680154Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[128x160x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8681492Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.8682805Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.8684120Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.8685433Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.8686760Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.8688090Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.8689431Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.8690760Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8692070Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.8693385Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.8694705Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.8696024Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.8697389Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:20.8698713Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.8700036Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:20.8701466Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8702907Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:20.8704225Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:20.8705548Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:20.8706951Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:20.8708509Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:20.8709910Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:20.8711397Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:20.8712740Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:20.8714071Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:21.0259233Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:21.0260637Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:21.0261977Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:21.0263311Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:21.0264895Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:21.0266230Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:21.0267580Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[64x64x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:21.0269356Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:21.0270850Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:21.0272506Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:21.0273871Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:21.0275453Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:21.0276923Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:21.0278710Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:21.0280436Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:21.0282143Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:21.0283500Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:21.0284865Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:21.0286240Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:21.0287577Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:21.0288925Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:21.0290275Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:21.0291850Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-True-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:21.0293444Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:21.0295149Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:21.0296630Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:21.0298000Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:21.0299370Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-True] SKIPPED 2025-09-09T14:44:21.0300788Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:21.0302549Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-True] PASSED 2025-09-09T14:44:21.0304327Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype0-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:21.0305855Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-True] SKIPPED 2025-09-09T14:44:21.0309004Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-True-False] SKIPPED 2025-09-09T14:44:21.0310350Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-True] SKIPPED 2025-09-09T14:44:21.0311796Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.DYNAMIC-False-False] SKIPPED 2025-09-09T14:44:21.0313148Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-True] XFAIL 2025-09-09T14:44:21.0314616Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-True-False] SKIPPED 2025-09-09T14:44:21.0316376Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-True] XFAIL 2025-09-09T14:44:21.0317968Z test/prototype/mx_formats/test_inference_workflow.py::test_inference_workflow_nvfp4[200x192x256-False-inpt_dtype1-NVFP4MMConfig.WEIGHT_ONLY-False-False] PASSED 2025-09-09T14:44:21.0319204Z test/prototype/mx_formats/test_kernels.py::test_fp32 SKIPPED (TODO d...) 2025-09-09T14:44:21.0319864Z test/prototype/mx_formats/test_kernels.py::test_bf16 SKIPPED (TODO d...) 2025-09-09T14:44:21.0320462Z test/prototype/mx_formats/test_kernels.py::test_fp16 PASSED 2025-09-09T14:44:21.0321063Z test/prototype/mx_formats/test_kernels.py::test_float8_e4m3fn PASSED 2025-09-09T14:44:21.0321681Z test/prototype/mx_formats/test_kernels.py::test_float8_e5m2 PASSED 2025-09-09T14:44:21.0322313Z test/prototype/mx_formats/test_kernels.py::test_float4_e2m1_table PASSED 2025-09-09T14:44:21.0323067Z test/prototype/mx_formats/test_kernels.py::test_float6_e3m2_table PASSED 2025-09-09T14:44:21.0323724Z test/prototype/mx_formats/test_kernels.py::test_float6_e2m3_table PASSED 2025-09-09T14:44:21.0324529Z test/prototype/mx_formats/test_kernels.py::test_fp4_0_0 PASSED 2025-09-09T14:44:21.0325294Z test/prototype/mx_formats/test_kernels.py::test_fp4_0_5 PASSED 2025-09-09T14:44:21.0325943Z test/prototype/mx_formats/test_kernels.py::test_fp4_1_0 PASSED 2025-09-09T14:44:21.0326519Z test/prototype/mx_formats/test_kernels.py::test_fp4_1_5 PASSED 2025-09-09T14:44:21.0327304Z test/prototype/mx_formats/test_kernels.py::test_fp4_2_0 PASSED 2025-09-09T14:44:21.0328092Z test/prototype/mx_formats/test_kernels.py::test_fp4_3_0 PASSED 2025-09-09T14:44:21.0328675Z test/prototype/mx_formats/test_kernels.py::test_fp4_4_0 PASSED 2025-09-09T14:44:21.0329257Z test/prototype/mx_formats/test_kernels.py::test_fp4_6_0 PASSED 2025-09-09T14:44:21.0329869Z test/prototype/mx_formats/test_kernels.py::test_fp4_pack_unpack PASSED 2025-09-09T14:44:21.0330538Z test/prototype/mx_formats/test_kernels.py::test_fp6_values[fp6_e2m3] PASSED 2025-09-09T14:44:21.0331204Z test/prototype/mx_formats/test_kernels.py::test_fp6_values[fp6_e3m2] PASSED 2025-09-09T14:44:21.0331920Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[29.0-31-cpu] PASSED 2025-09-09T14:44:21.0332667Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[29.0-31-cuda] PASSED 2025-09-09T14:44:31.7569486Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[26.0-30-cpu] PASSED 2025-09-09T14:44:31.7570270Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[26.0-30-cuda] PASSED 2025-09-09T14:44:31.7571013Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[0.1251-2-cpu] PASSED 2025-09-09T14:44:31.7571763Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[0.1251-2-cuda] PASSED 2025-09-09T14:44:31.7572806Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[0.0314-1-cpu] PASSED 2025-09-09T14:44:31.7573551Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[0.0314-1-cuda] PASSED 2025-09-09T14:44:31.7574542Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[0.03-0-cpu] PASSED 2025-09-09T14:44:31.7575277Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_rounding[0.03-0-cuda] PASSED 2025-09-09T14:44:31.7575985Z test/prototype/mx_formats/test_kernels.py::test_fp6_e2m3_pack_unpack PASSED 2025-09-09T14:44:31.7576651Z test/prototype/mx_formats/test_kernels.py::test_fp6_e3m2_pack_unpack PASSED 2025-09-09T14:44:31.7577376Z test/prototype/mx_formats/test_kernels.py::test_triton_mxfp8_dim1_randn[256-256] SKIPPED 2025-09-09T14:44:31.7578140Z test/prototype/mx_formats/test_kernels.py::test_triton_mxfp8_dim1_randn[256-2048] SKIPPED 2025-09-09T14:44:31.7578903Z test/prototype/mx_formats/test_kernels.py::test_triton_mxfp8_dim1_randn[2048-256] SKIPPED 2025-09-09T14:44:31.7579662Z test/prototype/mx_formats/test_kernels.py::test_triton_mxfp8_dim1_randn[2048-2048] SKIPPED 2025-09-09T14:44:31.7580360Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape0] PASSED 2025-09-09T14:44:31.7581012Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape1] PASSED 2025-09-09T14:44:31.7581650Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape2] PASSED 2025-09-09T14:44:31.7582293Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape3] PASSED 2025-09-09T14:44:31.7582936Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape4] PASSED 2025-09-09T14:44:31.7583623Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape5] PASSED 2025-09-09T14:44:31.7584267Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape6] PASSED 2025-09-09T14:44:31.7584907Z test/prototype/mx_formats/test_kernels.py::test_rearrange[shape7] PASSED 2025-09-09T14:44:31.7585804Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-32-32] SKIPPED 2025-09-09T14:44:31.7586888Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-32-64] SKIPPED 2025-09-09T14:44:31.7587968Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-32-2048] SKIPPED 2025-09-09T14:44:31.7589051Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-64-32] SKIPPED 2025-09-09T14:44:31.7590122Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-64-64] SKIPPED 2025-09-09T14:44:31.7591192Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-64-2048] SKIPPED 2025-09-09T14:44:31.7592284Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-2048-32] SKIPPED 2025-09-09T14:44:31.7593376Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-2048-64] SKIPPED 2025-09-09T14:44:31.7594518Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype0-2048-2048] SKIPPED 2025-09-09T14:44:31.7595607Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-32-32] SKIPPED 2025-09-09T14:44:31.7596798Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-32-64] SKIPPED 2025-09-09T14:44:31.7597877Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-32-2048] SKIPPED 2025-09-09T14:44:31.7599047Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-64-32] SKIPPED 2025-09-09T14:44:31.7600106Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-64-64] SKIPPED 2025-09-09T14:44:31.7601265Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-64-2048] SKIPPED 2025-09-09T14:44:31.7602339Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-2048-32] SKIPPED 2025-09-09T14:44:31.7603420Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-2048-64] SKIPPED 2025-09-09T14:44:31.7604556Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.FLOOR-input_dtype1-2048-2048] SKIPPED 2025-09-09T14:44:31.7605634Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-32-32] SKIPPED 2025-09-09T14:44:31.7606710Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-32-64] SKIPPED 2025-09-09T14:44:31.7607791Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-32-2048] SKIPPED 2025-09-09T14:44:31.7608854Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-64-32] SKIPPED 2025-09-09T14:44:31.7609913Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-64-64] SKIPPED 2025-09-09T14:44:31.7610982Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-64-2048] SKIPPED 2025-09-09T14:44:31.7612232Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-2048-32] SKIPPED 2025-09-09T14:44:31.7613323Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-2048-64] SKIPPED 2025-09-09T14:44:31.7614467Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype0-2048-2048] SKIPPED 2025-09-09T14:44:31.7615543Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-32-32] SKIPPED 2025-09-09T14:44:31.7616595Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-32-64] SKIPPED 2025-09-09T14:44:31.7617668Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-32-2048] SKIPPED 2025-09-09T14:44:31.7618738Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-64-32] SKIPPED 2025-09-09T14:44:31.7619803Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-64-64] SKIPPED 2025-09-09T14:44:31.7620880Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-64-2048] SKIPPED 2025-09-09T14:44:31.7621956Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-2048-32] SKIPPED 2025-09-09T14:44:31.7623027Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-2048-64] SKIPPED 2025-09-09T14:44:31.7624168Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_numerics[ScaleCalculationMode.RCEIL-input_dtype1-2048-2048] SKIPPED 2025-09-09T14:44:31.7625057Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim0_not_supported SKIPPED 2025-09-09T14:44:31.7625900Z test/prototype/mx_formats/test_kernels.py::test_cuda_mx_dim1_invalid_block_size SKIPPED 2025-09-09T14:44:31.7627051Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype0] PASSED 2025-09-09T14:44:31.7628617Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype1] PASSED 2025-09-09T14:44:31.7630098Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype2] PASSED 2025-09-09T14:44:31.7631570Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype3] PASSED 2025-09-09T14:44:31.7633039Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype4] PASSED 2025-09-09T14:44:31.7634519Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype0] PASSED 2025-09-09T14:44:31.9800897Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype1] PASSED 2025-09-09T14:44:31.9802461Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype2] PASSED 2025-09-09T14:44:31.9804231Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype3] PASSED 2025-09-09T14:44:31.9805830Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype4] PASSED 2025-09-09T14:44:31.9807544Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype0] PASSED 2025-09-09T14:44:31.9809175Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype1] PASSED 2025-09-09T14:44:31.9810674Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype2] PASSED 2025-09-09T14:44:31.9812276Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype3] PASSED 2025-09-09T14:44:31.9813992Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype4] PASSED 2025-09-09T14:44:31.9823183Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype0] PASSED 2025-09-09T14:44:31.9824926Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype1] PASSED 2025-09-09T14:44:31.9826544Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype2] PASSED 2025-09-09T14:44:31.9828043Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype3] PASSED 2025-09-09T14:44:31.9829607Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype4] PASSED 2025-09-09T14:44:31.9831429Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype0] PASSED 2025-09-09T14:44:31.9833172Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype1] PASSED 2025-09-09T14:44:31.9834794Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype2] PASSED 2025-09-09T14:44:31.9836376Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype3] PASSED 2025-09-09T14:44:31.9837860Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype4] PASSED 2025-09-09T14:44:31.9839476Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype0] PASSED 2025-09-09T14:44:31.9841078Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype1] PASSED 2025-09-09T14:44:31.9842680Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype2] PASSED 2025-09-09T14:44:31.9844319Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype3] PASSED 2025-09-09T14:44:31.9845804Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype4] PASSED 2025-09-09T14:44:31.9847298Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:31.9848840Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:31.9850491Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:31.9852032Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:31.9853707Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:31.9855259Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:31.9856777Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:31.9858372Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:31.9859970Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:31.9861458Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:31.9863068Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:31.9865599Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:31.9867100Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:31.9868592Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:31.9870085Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:31.9871584Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:31.9873316Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:31.9874919Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:31.9876470Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:31.9878015Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:31.9879698Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:31.9881196Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0931753Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0933456Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0934969Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0936474Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0937981Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0939475Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0940990Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0942490Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0944216Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0945811Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0947295Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0948763Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0950235Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0951716Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0953195Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0954681Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0956163Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0957717Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0959196Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0960669Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0962133Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0963602Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0965329Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0966805Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0968289Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0969760Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0971242Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0972720Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0974536Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0976130Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0977611Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0979074Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0980543Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0982024Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.0983502Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.0984984Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.0986466Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.0987936Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.0989422Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype0] PASSED 2025-09-09T14:44:32.0990893Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype1] PASSED 2025-09-09T14:44:32.0992353Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype2] PASSED 2025-09-09T14:44:32.0993826Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype3] PASSED 2025-09-09T14:44:32.0995344Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype4] PASSED 2025-09-09T14:44:32.0996854Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype0] PASSED 2025-09-09T14:44:32.0998331Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype1] PASSED 2025-09-09T14:44:32.0999797Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype2] PASSED 2025-09-09T14:44:32.2986572Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype3] PASSED 2025-09-09T14:44:32.2988102Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype4] PASSED 2025-09-09T14:44:32.2989967Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype0] PASSED 2025-09-09T14:44:32.2991562Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype1] PASSED 2025-09-09T14:44:32.2993036Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype2] PASSED 2025-09-09T14:44:32.2994506Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype3] PASSED 2025-09-09T14:44:32.2995961Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype4] PASSED 2025-09-09T14:44:32.2997538Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype0] PASSED 2025-09-09T14:44:32.2999018Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype1] PASSED 2025-09-09T14:44:32.3000482Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype2] PASSED 2025-09-09T14:44:32.3001949Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype3] PASSED 2025-09-09T14:44:32.3003415Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype4] PASSED 2025-09-09T14:44:32.3004882Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype0] PASSED 2025-09-09T14:44:32.3006353Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype1] PASSED 2025-09-09T14:44:32.3007812Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype2] PASSED 2025-09-09T14:44:32.3009278Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype3] PASSED 2025-09-09T14:44:32.3010743Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype4] PASSED 2025-09-09T14:44:32.3012213Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype0] PASSED 2025-09-09T14:44:32.3013693Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype1] PASSED 2025-09-09T14:44:32.3015214Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype2] PASSED 2025-09-09T14:44:32.3016679Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype3] PASSED 2025-09-09T14:44:32.3018147Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype4] PASSED 2025-09-09T14:44:32.3019729Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.3021328Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.3022818Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.3024351Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.3025828Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.3027321Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.3028814Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.3030299Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.3031793Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.3033275Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.3034771Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.3036324Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.3037803Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.3039288Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.3040776Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.3042261Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.3043761Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.3045298Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.3046783Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.3048280Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.3049861Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.3051422Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.3052906Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.3054387Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4284682Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4286222Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4287714Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4289196Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4290681Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4292164Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4293641Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4295158Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4296615Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4298063Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4299518Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4300985Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4302448Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4303917Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4305431Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4306893Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4308538Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4310107Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4311569Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4313026Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4314528Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4315999Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4317529Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4318989Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4320455Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4321917Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4323395Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4324891Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4326350Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4327793Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4329256Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4330725Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.4332191Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.4333662Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.4335134Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.4336594Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.4338154Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype0] PASSED 2025-09-09T14:44:32.4339695Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype1] PASSED 2025-09-09T14:44:32.4341161Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype2] PASSED 2025-09-09T14:44:32.4342626Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype3] PASSED 2025-09-09T14:44:32.4344111Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype4] PASSED 2025-09-09T14:44:32.4345610Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype0] PASSED 2025-09-09T14:44:32.4347091Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype1] PASSED 2025-09-09T14:44:32.4348558Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype2] PASSED 2025-09-09T14:44:32.4350027Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype3] PASSED 2025-09-09T14:44:32.4351501Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype4] PASSED 2025-09-09T14:44:32.6283154Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype0] PASSED 2025-09-09T14:44:32.6284708Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype1] PASSED 2025-09-09T14:44:32.6286181Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype2] PASSED 2025-09-09T14:44:32.6287648Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype3] PASSED 2025-09-09T14:44:32.6289119Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype4] PASSED 2025-09-09T14:44:32.6290602Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype0] PASSED 2025-09-09T14:44:32.6292079Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype1] PASSED 2025-09-09T14:44:32.6293562Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype2] PASSED 2025-09-09T14:44:32.6295042Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype3] PASSED 2025-09-09T14:44:32.6296508Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype4] PASSED 2025-09-09T14:44:32.6298181Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype0] PASSED 2025-09-09T14:44:32.6299772Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype1] PASSED 2025-09-09T14:44:32.6301241Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype2] PASSED 2025-09-09T14:44:32.6302720Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype3] PASSED 2025-09-09T14:44:32.6304250Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype4] PASSED 2025-09-09T14:44:32.6305727Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype0] PASSED 2025-09-09T14:44:32.6307207Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype1] PASSED 2025-09-09T14:44:32.6308691Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype2] PASSED 2025-09-09T14:44:32.6310163Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype3] PASSED 2025-09-09T14:44:32.6311634Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype4] PASSED 2025-09-09T14:44:32.6313120Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.6314652Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.6316136Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.6317710Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.6319187Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.6320687Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.6322175Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.6323668Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.6325158Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.6326644Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.6328225Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.6329793Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.6331271Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.6332753Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.6334291Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.6335775Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.6337267Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.6338756Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.6340240Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.6341733Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.6343236Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.6344775Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.6346265Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.6347752Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.6349227Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.6350728Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7673097Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7674705Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7676198Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7677773Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7679461Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7681048Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7682526Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7684052Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7685517Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7686995Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7688479Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7689947Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7691603Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7693071Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7694603Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7696079Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7697536Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7699008Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7700481Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7701951Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7703432Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7704915Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7706382Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7707853Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7709426Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7710964Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7712432Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7713896Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7715360Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7716882Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.7718368Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.7719837Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.7721305Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.7722775Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.EVEN-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.7724284Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype0] PASSED 2025-09-09T14:44:32.7725798Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype1] PASSED 2025-09-09T14:44:32.7727268Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype2] PASSED 2025-09-09T14:44:32.7728754Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype3] PASSED 2025-09-09T14:44:32.7730239Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-True-elem_dtype4] PASSED 2025-09-09T14:44:32.7731722Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype0] PASSED 2025-09-09T14:44:32.7733214Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype1] PASSED 2025-09-09T14:44:32.7734698Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype2] PASSED 2025-09-09T14:44:32.7736180Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype3] PASSED 2025-09-09T14:44:32.7737662Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape0-False-elem_dtype4] PASSED 2025-09-09T14:44:32.7739253Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype0] PASSED 2025-09-09T14:44:32.7740814Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype1] PASSED 2025-09-09T14:44:32.9355788Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype2] PASSED 2025-09-09T14:44:32.9357381Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype3] PASSED 2025-09-09T14:44:32.9358867Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-True-elem_dtype4] PASSED 2025-09-09T14:44:32.9360354Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype0] PASSED 2025-09-09T14:44:32.9361843Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype1] PASSED 2025-09-09T14:44:32.9363322Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype2] PASSED 2025-09-09T14:44:32.9364996Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype3] PASSED 2025-09-09T14:44:32.9366482Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape1-False-elem_dtype4] PASSED 2025-09-09T14:44:32.9367967Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype0] PASSED 2025-09-09T14:44:32.9369448Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype1] PASSED 2025-09-09T14:44:32.9370928Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype2] PASSED 2025-09-09T14:44:32.9372408Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype3] PASSED 2025-09-09T14:44:32.9373880Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-True-elem_dtype4] PASSED 2025-09-09T14:44:32.9375414Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype0] PASSED 2025-09-09T14:44:32.9376904Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype1] PASSED 2025-09-09T14:44:32.9378389Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype2] PASSED 2025-09-09T14:44:32.9379872Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype3] PASSED 2025-09-09T14:44:32.9381349Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-input_shape2-False-elem_dtype4] PASSED 2025-09-09T14:44:32.9383051Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.9384714Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.9386208Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.9387707Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.9389199Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.9390701Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.9392209Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.9393718Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.9395214Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.9396763Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.9398267Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.9399763Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.9401271Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.9402767Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.9404295Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.9405820Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.9407328Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.9408829Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:32.9410333Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:32.9411839Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:32.9413436Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:32.9415063Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:32.9416554Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:32.9418052Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:32.9419551Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:32.9421050Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:32.9422562Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:32.9424069Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2343589Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2345196Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2346871Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype0] SKIPPED 2025-09-09T14:44:48.2348602Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype1] SKIPPED 2025-09-09T14:44:48.2350169Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2351762Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2353250Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-True-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2354793Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype0] SKIPPED 2025-09-09T14:44:48.2356403Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype1] SKIPPED 2025-09-09T14:44:48.2358046Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2359549Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2361228Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape0-False-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2364209Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype0] SKIPPED 2025-09-09T14:44:48.2365885Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype1] SKIPPED 2025-09-09T14:44:48.2367589Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2369073Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2370629Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-True-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2372207Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype0] SKIPPED 2025-09-09T14:44:48.2373690Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype1] SKIPPED 2025-09-09T14:44:48.2375159Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2376632Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2378110Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape1-False-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2379681Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype0] SKIPPED 2025-09-09T14:44:48.2381281Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype1] SKIPPED 2025-09-09T14:44:48.2382758Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2384399Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2385881Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-True-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2387364Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype0] SKIPPED 2025-09-09T14:44:48.2388838Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype1] SKIPPED 2025-09-09T14:44:48.2390492Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype2] SKIPPED 2025-09-09T14:44:48.2391993Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype3] SKIPPED 2025-09-09T14:44:48.2393657Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_vs_hp[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-input_shape2-False-elem_dtype4] SKIPPED 2025-09-09T14:44:48.2395155Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_emulated_vs_real_gemm[mkn0-MXLinearRecipeName.MXFP8_CUBLAS] SKIPPED 2025-09-09T14:44:48.2396404Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_emulated_vs_real_gemm[mkn0-MXLinearRecipeName.MXFP4_CUTLASS] SKIPPED 2025-09-09T14:44:48.2397501Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_emulated_vs_real_gemm[mkn1-MXLinearRecipeName.MXFP8_CUBLAS] SKIPPED 2025-09-09T14:44:48.2398590Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_emulated_vs_real_gemm[mkn1-MXLinearRecipeName.MXFP4_CUTLASS] SKIPPED 2025-09-09T14:44:48.2399700Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_emulated_vs_real_gemm[mkn2-MXLinearRecipeName.MXFP8_CUBLAS] SKIPPED 2025-09-09T14:44:48.2400782Z test/prototype/mx_formats/test_mx_linear.py::test_linear_eager_emulated_vs_real_gemm[mkn2-MXLinearRecipeName.MXFP4_CUTLASS] SKIPPED 2025-09-09T14:44:48.2401688Z test/prototype/mx_formats/test_mx_linear.py::test_activation_checkpointing PASSED 2025-09-09T14:44:48.2402836Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:48.2404353Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:48.2413211Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_emulated-hp_dtype0] W0909 14:44:40.693370 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2414625Z W0909 14:44:40.697765 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2415314Z W0909 14:44:40.823016 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2415994Z W0909 14:44:40.827028 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2416668Z W0909 14:44:41.008126 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2417341Z W0909 14:44:41.012316 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2418017Z W0909 14:44:41.083362 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2418679Z W0909 14:44:41.085447 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:48.2419211Z PASSED 2025-09-09T14:44:48.2420122Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2645156Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2646729Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2648479Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2650161Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2651765Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2653692Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2655525Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_emulated-hp_dtype0] W0909 14:44:53.825066 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2656964Z W0909 14:44:53.829368 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2657715Z W0909 14:44:53.916803 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2658414Z W0909 14:44:53.920937 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2659086Z W0909 14:44:54.063856 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2659814Z W0909 14:44:54.068096 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2660582Z W0909 14:44:54.104998 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2661351Z W0909 14:44:54.107036 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:44:57.2661865Z PASSED 2025-09-09T14:44:57.2662780Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2664433Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2665924Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2667586Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2669084Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2670764Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2672324Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2673842Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2675369Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2676954Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2678451Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2679952Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2681456Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2683189Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2684804Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2686310Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2687801Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2689295Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2690784Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2692276Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2693776Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2695272Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2696754Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2698240Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2699785Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2701270Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2702743Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2704212Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2705701Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2707192Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2708671Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2710204Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2711682Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:44:57.2713153Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:44:57.2714719Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7534202Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7535754Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.FLOOR-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7537258Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7538755Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7540444Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_emulated-hp_dtype0] W0909 14:45:03.543421 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7541775Z W0909 14:45:03.547878 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7542453Z W0909 14:45:03.635829 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7543169Z W0909 14:45:03.640028 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7543826Z W0909 14:45:03.823020 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7544484Z W0909 14:45:03.827320 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7545143Z W0909 14:45:03.864457 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7545805Z W0909 14:45:03.866668 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7546309Z PASSED 2025-09-09T14:45:17.7547200Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7548688Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7550158Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7551638Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7553181Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7554658Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7556134Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7557890Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_emulated-hp_dtype0] W0909 14:45:14.058793 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7559179Z W0909 14:45:14.063003 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7560033Z W0909 14:45:14.148683 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7560695Z W0909 14:45:14.152810 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7561427Z W0909 14:45:14.303914 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7562089Z W0909 14:45:14.308042 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7562799Z W0909 14:45:14.344045 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7563465Z W0909 14:45:14.345942 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:17.7564256Z PASSED 2025-09-09T14:45:17.7565139Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7566624Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7568107Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7569567Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7571040Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7572548Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7574084Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7575594Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7577088Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7578582Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7580079Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7581578Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7583133Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7584625Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7586107Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7587602Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7589241Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7590830Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7592323Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7593856Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7595332Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7596891Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7598379Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:17.7599858Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:17.7601341Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4715004Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4716653Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4718148Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4719626Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4721105Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4722570Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4724044Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4725522Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4726984Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4728450Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4729915Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4731368Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.CEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4733190Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4734933Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4736598Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_emulated-hp_dtype0] W0909 14:45:24.738790 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4737927Z W0909 14:45:24.741957 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4738604Z W0909 14:45:24.797164 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4739267Z W0909 14:45:24.800117 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4739928Z W0909 14:45:24.906427 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4740594Z W0909 14:45:24.909397 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4741246Z W0909 14:45:24.940792 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4741913Z W0909 14:45:24.942719 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4742403Z PASSED 2025-09-09T14:45:36.4743309Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4744863Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4746357Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4747855Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4749347Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4750836Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4752333Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4753993Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_emulated-hp_dtype0] W0909 14:45:33.563286 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4755345Z W0909 14:45:33.566432 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4756013Z W0909 14:45:33.621481 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4756774Z W0909 14:45:33.624389 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4757440Z W0909 14:45:33.730213 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4758101Z W0909 14:45:33.733272 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4758853Z W0909 14:45:33.764146 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4759516Z W0909 14:45:33.766120 915 site-packages/torch/_inductor/utils.py:2237] [0/0] DeviceCopy in input program 2025-09-09T14:45:36.4760086Z PASSED 2025-09-09T14:45:36.4760978Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4762468Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4764297Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4766120Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4767602Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TORCH-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4769108Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4770633Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4772138Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4773652Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4775207Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4776709Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4778211Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4779714Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.4781219Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.4782724Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5363546Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5365703Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5367215Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5368715Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5370392Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5372009Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.TRITON-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5373511Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5375029Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5376539Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5378043Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5379542Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5381036Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5382515Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5384018Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-False-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5385531Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5387027Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5388530Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_emulated-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5390028Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_emulated-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5391507Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_cublas-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5393002Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp8_cublas-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5394507Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_cutlass-hp_dtype0] SKIPPED 2025-09-09T14:45:36.5396024Z test/prototype/mx_formats/test_mx_linear.py::test_linear_compile[ScaleCalculationMode.RCEIL-MXFP8Dim1CastKernelChoice.CUDA-True-mxfp4_cutlass-hp_dtype1] SKIPPED 2025-09-09T14:45:36.5397161Z test/prototype/mx_formats/test_mx_linear.py::test_filter_fn PASSED 2025-09-09T14:45:36.5397818Z test/prototype/mx_formats/test_mx_linear.py::test_training_print_str PASSED 2025-09-09T14:45:36.5398573Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-128x128x128] SKIPPED 2025-09-09T14:45:36.5399460Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-256x256x256] SKIPPED 2025-09-09T14:45:36.5400241Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-384x384x384] SKIPPED 2025-09-09T14:45:36.5401019Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-512x512x512] SKIPPED 2025-09-09T14:45:36.5401911Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-768x768x768] SKIPPED 2025-09-09T14:45:36.5402717Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-1024x1024x1024] SKIPPED 2025-09-09T14:45:36.5403528Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-8192x8192x8192] SKIPPED 2025-09-09T14:45:36.5404314Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-128x256x384] SKIPPED 2025-09-09T14:45:36.5405099Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-256x384x512] SKIPPED 2025-09-09T14:45:36.5405877Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-129x256x384] SKIPPED 2025-09-09T14:45:36.5406656Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp8-133x512x528] SKIPPED 2025-09-09T14:45:36.5407439Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-128x128x128] SKIPPED 2025-09-09T14:45:36.5408217Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-256x256x256] SKIPPED 2025-09-09T14:45:36.5409004Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-384x384x384] SKIPPED 2025-09-09T14:45:36.5409777Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-512x512x512] SKIPPED 2025-09-09T14:45:36.5410558Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-768x768x768] SKIPPED 2025-09-09T14:45:36.5411356Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-1024x1024x1024] SKIPPED 2025-09-09T14:45:36.5412155Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-8192x8192x8192] SKIPPED 2025-09-09T14:45:36.5412947Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-128x256x384] SKIPPED 2025-09-09T14:45:36.5413728Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-256x384x512] SKIPPED 2025-09-09T14:45:36.5414510Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-129x256x384] SKIPPED 2025-09-09T14:45:36.5415297Z test/prototype/mx_formats/test_mx_mm.py::test_matrix_multiplication[fp4-133x512x528] SKIPPED 2025-09-09T14:45:36.5416040Z test/prototype/mx_formats/test_mx_tensor.py::test_hello_world[elem_dtype0] PASSED 2025-09-09T14:45:36.5416760Z test/prototype/mx_formats/test_mx_tensor.py::test_hello_world[elem_dtype1] PASSED 2025-09-09T14:45:36.5417448Z test/prototype/mx_formats/test_mx_tensor.py::test_hello_world[fp6_e2m3] PASSED 2025-09-09T14:45:36.5418129Z test/prototype/mx_formats/test_mx_tensor.py::test_hello_world[fp6_e3m2] PASSED 2025-09-09T14:45:36.5418815Z test/prototype/mx_formats/test_mx_tensor.py::test_hello_world[elem_dtype4] PASSED 2025-09-09T14:45:36.5419685Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype0-ScaleCalculationMode.FLOOR] PASSED 2025-09-09T14:45:36.5420683Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype0-ScaleCalculationMode.RCEIL] PASSED 2025-09-09T14:45:36.5421663Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype0-ScaleCalculationMode.CEIL] PASSED 2025-09-09T14:45:36.5422651Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype0-ScaleCalculationMode.EVEN] PASSED 2025-09-09T14:45:36.5423642Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype1-ScaleCalculationMode.FLOOR] PASSED 2025-09-09T14:45:36.5424765Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype1-ScaleCalculationMode.RCEIL] PASSED 2025-09-09T14:45:36.5425754Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype1-ScaleCalculationMode.CEIL] PASSED 2025-09-09T14:45:36.5426809Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype1-ScaleCalculationMode.EVEN] PASSED 2025-09-09T14:45:36.5427784Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e2m3-ScaleCalculationMode.FLOOR] PASSED 2025-09-09T14:45:36.5428747Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e2m3-ScaleCalculationMode.RCEIL] PASSED 2025-09-09T14:45:36.5429699Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e2m3-ScaleCalculationMode.CEIL] PASSED 2025-09-09T14:45:36.5430650Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e2m3-ScaleCalculationMode.EVEN] PASSED 2025-09-09T14:45:52.2252575Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e3m2-ScaleCalculationMode.FLOOR] PASSED 2025-09-09T14:45:52.2253625Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e3m2-ScaleCalculationMode.RCEIL] PASSED 2025-09-09T14:45:52.2254625Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e3m2-ScaleCalculationMode.CEIL] PASSED 2025-09-09T14:45:52.2255581Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[fp6_e3m2-ScaleCalculationMode.EVEN] PASSED 2025-09-09T14:45:52.2256558Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype4-ScaleCalculationMode.FLOOR] PASSED 2025-09-09T14:45:52.2257553Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype4-ScaleCalculationMode.RCEIL] PASSED 2025-09-09T14:45:52.2258562Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype4-ScaleCalculationMode.CEIL] PASSED 2025-09-09T14:45:52.2259576Z test/prototype/mx_formats/test_mx_tensor.py::test_realistic_numerics[elem_dtype4-ScaleCalculationMode.EVEN] PASSED 2025-09-09T14:45:52.2260405Z test/prototype/mx_formats/test_mx_tensor.py::test_all_zeros[elem_dtype0] PASSED 2025-09-09T14:45:52.2261111Z test/prototype/mx_formats/test_mx_tensor.py::test_all_zeros[elem_dtype1] PASSED 2025-09-09T14:45:52.2261791Z test/prototype/mx_formats/test_mx_tensor.py::test_all_zeros[fp6_e2m3] PASSED 2025-09-09T14:45:52.2262463Z test/prototype/mx_formats/test_mx_tensor.py::test_all_zeros[fp6_e3m2] PASSED 2025-09-09T14:45:52.2263145Z test/prototype/mx_formats/test_mx_tensor.py::test_all_zeros[elem_dtype4] PASSED 2025-09-09T14:45:52.2264002Z test/prototype/mx_formats/test_mx_tensor.py::test_some_zeros[elem_dtype0] PASSED 2025-09-09T14:45:52.2264758Z test/prototype/mx_formats/test_mx_tensor.py::test_some_zeros[elem_dtype1] PASSED 2025-09-09T14:45:52.2265440Z test/prototype/mx_formats/test_mx_tensor.py::test_some_zeros[fp6_e2m3] PASSED 2025-09-09T14:45:52.2266108Z test/prototype/mx_formats/test_mx_tensor.py::test_some_zeros[fp6_e3m2] PASSED 2025-09-09T14:45:52.2266797Z test/prototype/mx_formats/test_mx_tensor.py::test_some_zeros[elem_dtype4] PASSED 2025-09-09T14:45:52.2267450Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_rceil SKIPPED 2025-09-09T14:45:52.2268141Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_in[elem_dtype0] PASSED 2025-09-09T14:45:52.2268870Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_in[elem_dtype1] PASSED 2025-09-09T14:45:52.2269601Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_in[fp6_e2m3] PASSED 2025-09-09T14:45:52.2270305Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_in[fp6_e3m2] PASSED 2025-09-09T14:45:52.2271027Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_in[elem_dtype4] PASSED 2025-09-09T14:45:52.2272060Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[False-elem_dtype0] PASSED 2025-09-09T14:45:52.2272834Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[False-elem_dtype1] PASSED 2025-09-09T14:45:52.2273767Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[False-fp6_e2m3] PASSED 2025-09-09T14:45:52.2274529Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[False-fp6_e3m2] PASSED 2025-09-09T14:45:52.2275298Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[False-elem_dtype4] PASSED 2025-09-09T14:45:52.2276073Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[True-elem_dtype0] SKIPPED 2025-09-09T14:45:52.2276952Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[True-elem_dtype1] SKIPPED 2025-09-09T14:45:52.2277713Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[True-fp6_e2m3] PASSED 2025-09-09T14:45:52.2278467Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[True-fp6_e3m2] PASSED 2025-09-09T14:45:52.2279234Z test/prototype/mx_formats/test_mx_tensor.py::test_exponent_nan_out[True-elem_dtype4] SKIPPED 2025-09-09T14:45:52.2279950Z test/prototype/mx_formats/test_mx_tensor.py::test_ranks[elem_dtype0] PASSED 2025-09-09T14:45:52.2280603Z test/prototype/mx_formats/test_mx_tensor.py::test_ranks[elem_dtype1] PASSED 2025-09-09T14:45:52.2281252Z test/prototype/mx_formats/test_mx_tensor.py::test_ranks[fp6_e2m3] PASSED 2025-09-09T14:45:52.2281883Z test/prototype/mx_formats/test_mx_tensor.py::test_ranks[fp6_e3m2] PASSED 2025-09-09T14:45:52.2282527Z test/prototype/mx_formats/test_mx_tensor.py::test_ranks[elem_dtype4] PASSED 2025-09-09T14:45:52.2283213Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[1-elem_dtype0] PASSED 2025-09-09T14:45:52.2283922Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[1-elem_dtype1] PASSED 2025-09-09T14:45:52.2284631Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[1-fp6_e2m3] SKIPPED 2025-09-09T14:45:52.2285326Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[1-fp6_e3m2] SKIPPED 2025-09-09T14:45:52.2286041Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[1-elem_dtype4] SKIPPED 2025-09-09T14:45:52.2286750Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[4-elem_dtype0] PASSED 2025-09-09T14:45:52.2287458Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[4-elem_dtype1] PASSED 2025-09-09T14:45:52.2288159Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[4-fp6_e2m3] PASSED 2025-09-09T14:45:52.2288838Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[4-fp6_e3m2] PASSED 2025-09-09T14:45:52.2289535Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[4-elem_dtype4] PASSED 2025-09-09T14:45:52.2290250Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[32-elem_dtype0] PASSED 2025-09-09T14:45:52.2290970Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[32-elem_dtype1] PASSED 2025-09-09T14:45:52.2291681Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[32-fp6_e2m3] PASSED 2025-09-09T14:45:52.2292374Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[32-fp6_e3m2] PASSED 2025-09-09T14:45:52.2293082Z test/prototype/mx_formats/test_mx_tensor.py::test_block_sizes[32-elem_dtype4] PASSED 2025-09-09T14:45:52.2293772Z test/prototype/mx_formats/test_mx_tensor.py::test_transpose[elem_dtype0] PASSED 2025-09-09T14:45:52.2294469Z test/prototype/mx_formats/test_mx_tensor.py::test_transpose[elem_dtype1] PASSED 2025-09-09T14:45:52.2295182Z test/prototype/mx_formats/test_mx_tensor.py::test_transpose[fp6_e2m3] PASSED 2025-09-09T14:45:52.2295937Z test/prototype/mx_formats/test_mx_tensor.py::test_transpose[fp6_e3m2] PASSED 2025-09-09T14:45:52.2296609Z test/prototype/mx_formats/test_mx_tensor.py::test_transpose[elem_dtype4] PASSED 2025-09-09T14:45:52.2297268Z test/prototype/mx_formats/test_mx_tensor.py::test_view[elem_dtype0] PASSED 2025-09-09T14:45:52.2298001Z test/prototype/mx_formats/test_mx_tensor.py::test_view[elem_dtype1] PASSED 2025-09-09T14:45:52.2298641Z test/prototype/mx_formats/test_mx_tensor.py::test_view[fp6_e2m3] PASSED 2025-09-09T14:45:52.2299281Z test/prototype/mx_formats/test_mx_tensor.py::test_view[fp6_e3m2] PASSED 2025-09-09T14:45:52.2299915Z test/prototype/mx_formats/test_mx_tensor.py::test_view[elem_dtype4] PASSED 2025-09-09T14:45:52.2300598Z test/prototype/mx_formats/test_mx_tensor.py::test_fp6_packing[False-fp6_e2m3] PASSED 2025-09-09T14:45:52.2301320Z test/prototype/mx_formats/test_mx_tensor.py::test_fp6_packing[False-fp6_e3m2] PASSED 2025-09-09T14:45:52.2302034Z test/prototype/mx_formats/test_mx_tensor.py::test_fp6_packing[True-fp6_e2m3] PASSED 2025-09-09T14:45:52.2302746Z test/prototype/mx_formats/test_mx_tensor.py::test_fp6_packing[True-fp6_e3m2] PASSED 2025-09-09T14:45:52.2303602Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype0-elem_dtype0] SKIPPED 2025-09-09T14:45:52.2304557Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype0-elem_dtype1] SKIPPED 2025-09-09T14:45:52.2305492Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype0-fp6_e2m3] PASSED 2025-09-09T14:45:52.2306400Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype0-fp6_e3m2] PASSED 2025-09-09T14:45:52.2307332Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype0-elem_dtype4] PASSED 2025-09-09T14:45:52.2308291Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype1-elem_dtype0] SKIPPED 2025-09-09T14:45:52.2309237Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype1-elem_dtype1] SKIPPED 2025-09-09T14:45:52.2310180Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype1-fp6_e2m3] PASSED 2025-09-09T14:45:52.2311093Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype1-fp6_e3m2] PASSED 2025-09-09T14:45:52.2312017Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[False-hp_dtype1-elem_dtype4] PASSED 2025-09-09T14:45:52.2312963Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype0-elem_dtype0] SKIPPED 2025-09-09T14:45:52.2313901Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype0-elem_dtype1] SKIPPED 2025-09-09T14:45:52.2314832Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype0-fp6_e2m3] PASSED 2025-09-09T14:45:56.6061080Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype0-fp6_e3m2] PASSED 2025-09-09T14:45:56.6062035Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype0-elem_dtype4] PASSED 2025-09-09T14:45:56.6062975Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype1-elem_dtype0] SKIPPED 2025-09-09T14:45:56.6064283Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype1-elem_dtype1] SKIPPED 2025-09-09T14:45:56.6065573Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype1-fp6_e2m3] PASSED 2025-09-09T14:45:56.6066695Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype1-fp6_e3m2] PASSED 2025-09-09T14:45:56.6067911Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_from_mx_compile_numerics[True-hp_dtype1-elem_dtype4] PASSED 2025-09-09T14:45:56.6068870Z test/prototype/mx_formats/test_mx_tensor.py::test_to_mx_inductor_single_kernel SKIPPED 2025-09-09T14:45:56.6069662Z test/prototype/mx_formats/test_mx_tensor.py::test_cast_to_float8_e4m3fn_saturation_behavior SKIPPED 2025-09-09T14:45:56.6070607Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_reconstruction[dtype0-shape0-False] PASSED 2025-09-09T14:45:56.6071616Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_reconstruction[dtype1-shape1-False] PASSED 2025-09-09T14:45:56.6072622Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_reconstruction[dtype2-shape2-False] PASSED 2025-09-09T14:45:56.6073619Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_reconstruction[dtype3-shape3-True] PASSED 2025-09-09T14:45:56.6074660Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[False-shape0] PASSED 2025-09-09T14:45:56.6075712Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[False-shape1] PASSED 2025-09-09T14:45:56.6076768Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[False-shape2] PASSED 2025-09-09T14:45:56.6077689Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[False-shape3] PASSED 2025-09-09T14:45:56.6078545Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[False-shape4] PASSED 2025-09-09T14:45:56.6079406Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[False-shape5] PASSED 2025-09-09T14:45:56.6080266Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[True-shape0] PASSED 2025-09-09T14:45:56.6081118Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[True-shape1] PASSED 2025-09-09T14:45:56.6081976Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[True-shape2] PASSED 2025-09-09T14:45:56.6082824Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[True-shape3] PASSED 2025-09-09T14:45:56.6083701Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[True-shape4] PASSED 2025-09-09T14:45:56.6084555Z test/prototype/mx_formats/test_mx_tensor.py::test_to_blocked_from_blocked_roundtrip[True-shape5] PASSED 2025-09-09T14:45:56.6085425Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape0-False] PASSED 2025-09-09T14:45:56.6086302Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape0-True] PASSED 2025-09-09T14:45:56.6093992Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape1-False] PASSED 2025-09-09T14:45:56.6095080Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape1-True] PASSED 2025-09-09T14:45:56.6096162Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape2-False] PASSED 2025-09-09T14:45:56.6097055Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape2-True] PASSED 2025-09-09T14:45:56.6097941Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape3-False] PASSED 2025-09-09T14:45:56.6098820Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_construction[shape3-True] PASSED 2025-09-09T14:45:56.6099704Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_rows[0:128]] PASSED 2025-09-09T14:45:56.6100704Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_rows[128:256]] PASSED 2025-09-09T14:45:56.6101582Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:64]] PASSED 2025-09-09T14:45:56.6102581Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[64:128]] PASSED 2025-09-09T14:45:56.6103506Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:128]_full_width] PASSED 2025-09-09T14:45:56.6104485Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:2048]_tp_first_half] PASSED 2025-09-09T14:45:56.6105485Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[2048:4096]_tp_second_half] PASSED 2025-09-09T14:45:56.6106454Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:1024]_quarter] PASSED 2025-09-09T14:45:56.6107450Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[1024:2048]_quarter] PASSED 2025-09-09T14:45:56.6108421Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_row_end] PASSED 2025-09-09T14:45:56.6109370Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_row_start] PASSED 2025-09-09T14:45:56.6110317Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_col_32] PASSED 2025-09-09T14:45:56.6111254Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_col_start] PASSED 2025-09-09T14:45:56.6112202Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_col_end] PASSED 2025-09-09T14:45:56.6113112Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[odd_start] PASSED 2025-09-09T14:45:56.6113981Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[odd_end] PASSED 2025-09-09T14:45:56.6114810Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_view_semantics PASSED 2025-09-09T14:45:56.6115599Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_serialization PASSED 2025-09-09T14:45:56.6116493Z test/prototype/mx_formats/test_mx_tensor.py::test_nvfp4_swizzled_scales_get_scales_method PASSED 2025-09-09T14:45:56.6117414Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.6118391Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.6119369Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.6120366Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.6121343Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.6122323Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.6123294Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.6124279Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.6125267Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.6126347Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.6127339Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.6128412Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.6129405Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.6130393Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.6131373Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.6132357Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.6372898Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.6373958Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.6374966Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.6375965Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.6376959Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.6378000Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.6379005Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.6379993Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.6381006Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.6382011Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.6383005Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.6383999Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.6384983Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.6385980Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.6386971Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.6388020Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.6389012Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.6390007Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.6390990Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.6392189Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.6393173Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.6394278Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.6395277Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.6396257Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.6397321Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.6398326Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.6399319Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.6400316Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.6401307Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.6402305Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.6403301Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.6404291Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.6405283Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.6406265Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.6407263Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.6408256Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.6409251Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.6410248Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.6411231Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.6412228Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.6413239Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.6414232Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.6415233Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.6416239Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.6417238Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.6418377Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.6419370Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.6420448Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.6421445Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.6422434Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.6423440Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.6424432Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.6425436Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.6426434Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.6427430Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.6428425Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.6429419Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.6430412Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.6431416Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.6432407Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.6433408Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.6434406Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.6435393Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.6683912Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.6684949Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.6685953Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.6686956Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.6688010Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.6689008Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.6690006Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.6690988Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.6692177Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.6693177Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.6694274Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.6695275Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.6696267Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.6697284Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.6698346Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.6699361Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.6700376Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.6701383Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.6702374Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.6703365Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.6704342Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.6705332Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.6706320Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.6707309Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.6708295Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.6709271Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.6710247Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.6711237Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.6712220Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.6713223Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.6714218Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.6715205Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.6716193Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.6717262Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.6718423Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.6719415Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.6720476Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.6721468Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.6722450Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.6723428Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.6724418Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.6725399Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.6726392Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.6727388Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.6728427Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.6729408Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.6730387Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.6731376Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.6732356Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.6733341Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.6734329Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.6735314Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.6736292Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.6737280Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.6738308Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.6739303Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.6740288Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.6741268Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.6742249Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.6743225Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.6744304Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.6745301Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.6746367Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.7174977Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.7176030Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.7177027Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.7178099Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.7179109Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.7180118Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.7181134Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.7182124Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.7183134Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.7184142Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.7185135Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.7186141Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.7187154Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.7188154Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.7189163Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.7190172Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.7191194Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.7192202Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.7193204Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.7194219Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.7195228Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.7196241Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.7197331Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.7198555Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.7199673Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.7200672Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.7201678Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.7202682Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.7203682Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.7204697Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.7205702Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.7206724Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.7207785Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.7208780Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.7209777Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.7210778Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.7211773Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.7212778Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.7213766Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.7214754Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.7215745Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.7216725Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.7217726Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.7218713Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.7219717Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.7220718Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.7221701Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.7222698Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.7223696Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.7224782Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.7225863Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.7226863Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.7227915Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.7228918Z test/prototype/mx_formats/test_mx_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.7229859Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_reconstruction[dtype0-shape0-False] PASSED 2025-09-09T14:45:56.7230747Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_reconstruction[dtype1-shape1-False] PASSED 2025-09-09T14:45:56.7231619Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_reconstruction[dtype2-shape2-False] PASSED 2025-09-09T14:45:56.7232499Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_reconstruction[dtype3-shape3-True] PASSED 2025-09-09T14:45:56.7233400Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape0-False] PASSED 2025-09-09T14:45:56.7234318Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape0-True] PASSED 2025-09-09T14:45:56.7235235Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape1-False] PASSED 2025-09-09T14:45:56.7236146Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape1-True] PASSED 2025-09-09T14:45:56.7237133Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape2-False] PASSED 2025-09-09T14:45:56.7238130Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape2-True] PASSED 2025-09-09T14:45:56.7797039Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape3-False] PASSED 2025-09-09T14:45:56.7798051Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_construction[shape3-True] PASSED 2025-09-09T14:45:56.7798975Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_rows[0:128]] PASSED 2025-09-09T14:45:56.7799891Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_rows[128:256]] PASSED 2025-09-09T14:45:56.7800808Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:64]] PASSED 2025-09-09T14:45:56.7801731Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[64:128]] PASSED 2025-09-09T14:45:56.7802686Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:128]_full_width] PASSED 2025-09-09T14:45:56.7803709Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:2048]_tp_first_half] PASSED 2025-09-09T14:45:56.7804743Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[2048:4096]_tp_second_half] PASSED 2025-09-09T14:45:56.7805754Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[0:1024]_quarter] PASSED 2025-09-09T14:45:56.7806762Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing[slice_cols[1024:2048]_quarter] PASSED 2025-09-09T14:45:56.7807807Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_row_end] PASSED 2025-09-09T14:45:56.7809005Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_row_start] PASSED 2025-09-09T14:45:56.7810096Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_col_32] PASSED 2025-09-09T14:45:56.7811084Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_col_start] PASSED 2025-09-09T14:45:56.7812074Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[misaligned_col_end] PASSED 2025-09-09T14:45:56.7813015Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[odd_start] PASSED 2025-09-09T14:45:56.7813928Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_slicing_errors[odd_end] PASSED 2025-09-09T14:45:56.7814794Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_view_semantics PASSED 2025-09-09T14:45:56.7815621Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_serialization PASSED 2025-09-09T14:45:56.7816478Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_swizzled_scales_get_scales_method PASSED 2025-09-09T14:45:56.7817437Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.7818485Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.7819524Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.7820556Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.7821604Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.7822627Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.7823657Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.7824689Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.7825710Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.7826738Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.7827770Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.7828804Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.7829839Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.7830860Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.7831890Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.7832928Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.7833948Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.7836187Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.7837289Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.7838402Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.7839433Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.7840452Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.7841475Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.7842512Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.7843535Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.7844573Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.7845595Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.7846622Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.7847647Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.7848663Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.7849687Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.7850715Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.7851731Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.7852749Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.7853761Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.7854775Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.7855801Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.7856815Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.7857847Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.7858863Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.7859880Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.7860892Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.8092056Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.8093266Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.8094410Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.8095445Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.8096475Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.8097554Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.8098573Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-block_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.8099611Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.8100637Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.8101666Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.8102698Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.8103728Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.8104762Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.8105792Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.8106820Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.8107862Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.8108892Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.8109933Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.8110974Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.8112004Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.8113050Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.8114096Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.8115123Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.8116161Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.8117280Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.8118336Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.8119475Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.8120619Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.8121666Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.8122702Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.8123744Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.8124804Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.8125849Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.8126902Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.8127959Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.8128997Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.8130030Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.8131063Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.8132106Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.8133155Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.8134191Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.8135223Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.8136261Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.8137287Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.8138319Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.8139362Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.8140406Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.8141443Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.8142468Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.8143508Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.8144543Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.8145680Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.8146735Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.8147855Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.8148900Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.8149942Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[fp32-tensor_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.8150967Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.8151999Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.8153020Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.8154058Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.8155086Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.8390315Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.8391516Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.8392535Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.8393557Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.8394580Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.8395611Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.8396745Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.8397759Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.8398788Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.8406662Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.8407773Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.8408817Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.8409860Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.8410899Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.8411927Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.8412949Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.8414181Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.8415322Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.8416353Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.8417381Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.8418454Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.8419479Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.8420511Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.8421531Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.8422551Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.8423562Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.8424580Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.8425600Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.8426610Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.8427645Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.8428685Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.8429696Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.8430706Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.8431717Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.8432739Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.8433756Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.8434768Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.8435781Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.8436880Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.8437954Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.8438985Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.8440106Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.8441126Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.8442224Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-block_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.8443242Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M128] SKIPPED 2025-09-09T14:45:56.8444261Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M256] SKIPPED 2025-09-09T14:45:56.8445284Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M512] SKIPPED 2025-09-09T14:45:56.8446314Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M1024] SKIPPED 2025-09-09T14:45:56.8447349Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M100] SKIPPED 2025-09-09T14:45:56.8448370Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M200] SKIPPED 2025-09-09T14:45:56.8449396Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N64-M384] SKIPPED 2025-09-09T14:45:56.8450427Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M128] SKIPPED 2025-09-09T14:45:56.8451453Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M256] SKIPPED 2025-09-09T14:45:56.8452488Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M512] SKIPPED 2025-09-09T14:45:56.8453534Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M1024] SKIPPED 2025-09-09T14:45:56.8454575Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M100] SKIPPED 2025-09-09T14:45:56.8455604Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M200] SKIPPED 2025-09-09T14:45:56.8456629Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N128-M384] SKIPPED 2025-09-09T14:45:56.8457710Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M128] SKIPPED 2025-09-09T14:45:56.8458737Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M256] SKIPPED 2025-09-09T14:45:56.8459774Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M512] SKIPPED 2025-09-09T14:45:56.8460809Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M1024] SKIPPED 2025-09-09T14:45:56.8657205Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M100] SKIPPED 2025-09-09T14:45:56.8658313Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M200] SKIPPED 2025-09-09T14:45:56.8659345Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N256-M384] SKIPPED 2025-09-09T14:45:56.8660368Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M128] SKIPPED 2025-09-09T14:45:56.8661394Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M256] SKIPPED 2025-09-09T14:45:56.8662589Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M512] SKIPPED 2025-09-09T14:45:56.8663880Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M1024] SKIPPED 2025-09-09T14:45:56.8664927Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M100] SKIPPED 2025-09-09T14:45:56.8665962Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M200] SKIPPED 2025-09-09T14:45:56.8666986Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N512-M384] SKIPPED 2025-09-09T14:45:56.8668066Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M128] SKIPPED 2025-09-09T14:45:56.8669088Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M256] SKIPPED 2025-09-09T14:45:56.8670120Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M512] SKIPPED 2025-09-09T14:45:56.8671146Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M1024] SKIPPED 2025-09-09T14:45:56.8672166Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M100] SKIPPED 2025-09-09T14:45:56.8673196Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M200] SKIPPED 2025-09-09T14:45:56.8674215Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N32-M384] SKIPPED 2025-09-09T14:45:56.8675241Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M128] SKIPPED 2025-09-09T14:45:56.8676258Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M256] SKIPPED 2025-09-09T14:45:56.8677367Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M512] SKIPPED 2025-09-09T14:45:56.8678402Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M1024] SKIPPED 2025-09-09T14:45:56.8679426Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M100] SKIPPED 2025-09-09T14:45:56.8680444Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M200] SKIPPED 2025-09-09T14:45:56.8681468Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N96-M384] SKIPPED 2025-09-09T14:45:56.8682499Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M128] SKIPPED 2025-09-09T14:45:56.8683537Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M256] SKIPPED 2025-09-09T14:45:56.8684563Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M512] SKIPPED 2025-09-09T14:45:56.8685601Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M1024] SKIPPED 2025-09-09T14:45:56.8686636Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M100] SKIPPED 2025-09-09T14:45:56.8687656Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M200] SKIPPED 2025-09-09T14:45:56.8688824Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_triton_nvfp4_quantize_equivalence[bf16-tensor_scale-N160-M384] SKIPPED 2025-09-09T14:45:56.8690106Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8691410Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8692720Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8694043Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8695353Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8696659Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8698035Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8699362Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8700674Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8701966Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8703272Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8704594Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8705898Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8707200Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8708565Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8709902Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8711226Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8712527Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8713836Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8715166Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8716676Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8717984Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8719389Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8720739Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8896206Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8897586Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8898943Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8900281Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8901622Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8902940Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8904259Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8905613Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8906940Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8908301Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8909621Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8910945Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8912276Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8913595Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8914923Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8916261Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8917663Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8918958Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8920439Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8921878Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8923191Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8924514Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8925835Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8927183Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8928519Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8929823Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8931149Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8932492Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8933818Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8935146Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8936483Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8937828Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8939158Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8940475Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8941809Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8943160Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8944487Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8945812Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8947154Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8948595Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[256x128x512-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8949992Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8951296Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8952605Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8953928Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8955246Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.8956646Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.8957978Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.8959324Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.8960640Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9140265Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9142946Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9145562Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9147850Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9149148Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9150459Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9151790Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9153095Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9154405Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9155722Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9157082Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9158567Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9160015Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9161343Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9162676Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9164289Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9165597Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9166916Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9168252Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9169562Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9170880Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9172199Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9173540Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[157x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9174861Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9176150Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9177479Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9178813Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9180125Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9181425Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9182732Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9184061Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9185362Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9186779Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9188078Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9189511Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9190821Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9192119Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9193432Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9194756Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9196067Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9197441Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9198751Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9200076Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9201401Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9202711Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9204036Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9205359Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9206678Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9207982Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9209293Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9383121Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9384473Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9385783Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9387118Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9388673Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x96x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9390101Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9391409Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9392725Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9394045Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9395372Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9396773Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9398106Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9399442Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9400770Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9402073Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9403391Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9404708Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9406020Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9407340Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9408701Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9410047Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9411377Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9412676Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9413992Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9415320Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9416748Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9418142Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9419470Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9420816Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9422145Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9423452Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9424778Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9426120Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9427441Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9428809Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9430144Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9431476Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[128x160x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9432793Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9434076Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9435370Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9436748Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9438102Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9439398Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9440708Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9442020Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9443316Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9444695Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9446089Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9447392Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9448689Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9619628Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9621084Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9622414Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9623719Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9625007Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9626299Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9627627Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9628958Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9630249Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9631563Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9632882Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9634176Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9635477Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9636829Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9638134Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9639437Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9640736Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9642040Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9643539Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[64x64x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9644959Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9646256Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9647567Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9648883Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9650204Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9651524Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9652845Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9654177Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9655492Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9656780Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9658094Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9659418Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9660734Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9662040Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9663364Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9664933Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-True-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9666255Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9667561Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9668880Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9670220Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9671682Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9673111Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9674450Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9675789Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype0-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9677170Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9678484Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-True-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:45:56.9679807Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:45:56.9681144Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-True-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:45:56.9682474Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-True] SKIPPED 2025-09-09T14:45:56.9683785Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-False-False-NVFP4MMConfig.DYNAMIC-False] SKIPPED 2025-09-09T14:50:25.1625000Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-True] SKIPPED 2025-09-09T14:50:25.1628143Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_matmul_with_amax[200x192x256-False-inpt_dtype1-False-False-NVFP4MMConfig.WEIGHT_ONLY-False] SKIPPED 2025-09-09T14:50:25.1629189Z test/prototype/mx_formats/test_nvfp4_tensor.py::test_nvfp4_to_copy PASSED 2025-09-09T14:50:25.1629990Z test/prototype/safetensors/test_safetensors_support.py::TestSafeTensors::test_safetensors SKIPPED 2025-09-09T14:50:25.1630920Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_metadata_torchao SKIPPED 2025-09-09T14:50:25.1631939Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata0 SKIPPED 2025-09-09T14:50:25.1632998Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata1 SKIPPED 2025-09-09T14:50:25.1634075Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata2 SKIPPED 2025-09-09T14:50:25.1635130Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata3 SKIPPED 2025-09-09T14:50:25.1636291Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata4 SKIPPED 2025-09-09T14:50:25.1637346Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata5 SKIPPED 2025-09-09T14:50:25.1638404Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata6 SKIPPED 2025-09-09T14:50:25.1639462Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata7 SKIPPED 2025-09-09T14:50:25.1642597Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata8 SKIPPED 2025-09-09T14:50:25.1643663Z test/prototype/safetensors/test_safetensors_utils.py::TestSafeTensorsUtils::test_not_metadata_torchao_metadata9 SKIPPED 2025-09-09T14:50:25.1644692Z test/prototype/test_awq.py::TestAWQ::test_awq_config SKIPPED (need t...) 2025-09-09T14:50:25.1645322Z test/prototype/test_awq.py::TestAWQ::test_awq_functionality SKIPPED 2025-09-09T14:50:25.1645947Z test/prototype/test_awq.py::TestAWQ::test_awq_loading SKIPPED (need ...) 2025-09-09T14:50:25.1646567Z test/prototype/test_awq.py::TestAWQ::test_awq_loading_vllm SKIPPED (...) 2025-09-09T14:50:25.1647237Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-2-512-128] Relative Error: 0.052430 2025-09-09T14:50:25.1647793Z PASSED 2025-09-09T14:50:25.1648275Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-3-2048-2048] Relative Error: 0.052625 2025-09-09T14:50:25.1648837Z PASSED 2025-09-09T14:50:25.1649308Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-4-3584-640] Relative Error: 0.052573 2025-09-09T14:50:25.1649860Z PASSED 2025-09-09T14:50:25.1650342Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-13-8704-8576] Relative Error: 0.052637 2025-09-09T14:50:25.1650907Z PASSED 2025-09-09T14:50:25.1651381Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-26-18944-1664] Relative Error: 0.052638 2025-09-09T14:50:25.1651949Z PASSED 2025-09-09T14:50:25.1652418Z test/prototype/test_blockwise_triton.py::test_blockwise_quant_dequant[dtype0-67-6656-1408] Relative Error: 0.052663 2025-09-09T14:50:25.1652976Z PASSED 2025-09-09T14:50:25.1653425Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-2-512-128] Relative Error: 0.076219 2025-09-09T14:50:25.1653957Z PASSED 2025-09-09T14:50:25.1654415Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-3-2048-2048] Relative Error: 0.073036 2025-09-09T14:50:25.1654946Z PASSED 2025-09-09T14:50:25.1655391Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-4-3584-640] Relative Error: 0.072511 2025-09-09T14:50:25.1655926Z PASSED 2025-09-09T14:50:25.1656381Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-13-8704-8576] Relative Error: 0.073267 2025-09-09T14:50:25.1656925Z PASSED 2025-09-09T14:50:25.1657376Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-26-18944-1664] Relative Error: 0.073289 2025-09-09T14:50:25.1657923Z PASSED 2025-09-09T14:50:25.1658373Z test/prototype/test_blockwise_triton.py::test_blockwise_fp8_gemm[dtype0-67-6656-1408] Relative Error: 0.073508 2025-09-09T14:50:25.1658910Z PASSED 2025-09-09T14:50:25.1659483Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_choose_qparams_codebook SKIPPED 2025-09-09T14:50:25.1660468Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_choose_qparams_codebook_row_grouping SKIPPED 2025-09-09T14:50:25.1661481Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float SKIPPED 2025-09-09T14:50:25.1662495Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float2 SKIPPED 2025-09-09T14:50:25.1663572Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float_row_grouping SKIPPED 2025-09-09T14:50:25.1664689Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_export SKIPPED 2025-09-09T14:50:25.1665503Z test/prototype/test_codebook_coreml.py::TestCodebookQuantization::test_quantize_api SKIPPED 2025-09-09T14:50:25.1666376Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_choose_qparams_codebook PASSED 2025-09-09T14:50:25.1667455Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float PASSED 2025-09-09T14:50:25.1668564Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_codebook_quantized_tensor_from_float2 PASSED 2025-09-09T14:50:25.1669465Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_quantize_api PASSED 2025-09-09T14:50:25.1670342Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1671277Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity0-dtype1] SKIPPED 2025-09-09T14:50:25.1672199Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity1-dtype0] SKIPPED 2025-09-09T14:50:25.1673131Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-1-granularity1-dtype1] SKIPPED 2025-09-09T14:50:25.1674072Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1674997Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity0-dtype1] SKIPPED 2025-09-09T14:50:25.1675994Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity1-dtype0] SKIPPED 2025-09-09T14:50:25.1676915Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-2-granularity1-dtype1] SKIPPED 2025-09-09T14:50:25.1677839Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1678769Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity0-dtype1] SKIPPED 2025-09-09T14:50:25.1679689Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity1-dtype0] SKIPPED 2025-09-09T14:50:25.1680626Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-3-granularity1-dtype1] SKIPPED 2025-09-09T14:50:25.1681551Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1682493Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity0-dtype1] SKIPPED 2025-09-09T14:50:25.1683424Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity1-dtype0] SKIPPED 2025-09-09T14:50:25.1684346Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim0-4-granularity1-dtype1] SKIPPED 2025-09-09T14:50:25.1685276Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1686200Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity0-dtype1] SKIPPED 2025-09-09T14:50:25.1687133Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity1-dtype0] SKIPPED 2025-09-09T14:50:25.1688065Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-1-granularity1-dtype1] SKIPPED 2025-09-09T14:50:25.1688984Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1689913Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity0-dtype1] SKIPPED 2025-09-09T14:50:25.1690831Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity1-dtype0] SKIPPED 2025-09-09T14:50:25.1691758Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-2-granularity1-dtype1] SKIPPED 2025-09-09T14:50:25.1692685Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity0-dtype0] SKIPPED 2025-09-09T14:50:25.1693694Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8034087Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8035326Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-3-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8036352Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8037277Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8038207Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8039134Z test/prototype/test_dynamic_activation_lut.py::test_parq_conversion[lead_dim1-4-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8040037Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8040896Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8041753Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8042615Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-1-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8043473Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8044322Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8045175Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8046034Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-2-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8046894Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8047760Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8048618Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8049482Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-3-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8050336Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8051199Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8052065Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8052919Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim0-4-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8053781Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8054638Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8055498Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8056359Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-1-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8057210Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8058069Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8059118Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8059976Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-2-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8060911Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8061768Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8062625Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8063477Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-3-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8064614Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity0-dtype0] SKIPPED 2025-09-09T14:50:31.8065480Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity0-dtype1] SKIPPED 2025-09-09T14:50:31.8066330Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity1-dtype0] SKIPPED 2025-09-09T14:50:31.8067192Z test/prototype/test_dynamic_activation_lut.py::test_export[lead_dim1-4-granularity1-dtype1] SKIPPED 2025-09-09T14:50:31.8068013Z test/prototype/test_gguf_quant.py::TestGGUFQuantization::test_choose_qparams_gguf PASSED 2025-09-09T14:50:31.8068854Z test/prototype/test_gguf_quant.py::TestGGUFQuantization::test_gguf_quantized_tensor_from_float PASSED 2025-09-09T14:50:31.8069651Z test/prototype/test_gguf_quant.py::TestGGUFQuantization::test_quantize_api PASSED 2025-09-09T14:50:31.8070461Z test/prototype/test_mixed_precision.py::TestWeightOnlyQuantNaive::test_quantization_intNwo PASSED 2025-09-09T14:50:31.8071260Z test/prototype/test_parametrization.py::TestFakeSparsity::test_jit_trace PASSED 2025-09-09T14:50:31.8072008Z test/prototype/test_parametrization.py::TestFakeSparsity::test_masking_logic PASSED 2025-09-09T14:50:31.8072800Z test/prototype/test_parametrization.py::TestFakeSparsity::test_state_dict_preserved PASSED 2025-09-09T14:50:31.8073625Z test/prototype/test_parametrization.py::TestFakeSparsity::test_weights_parametrized PASSED 2025-09-09T14:50:31.8074371Z test/prototype/test_paretoq.py::TestParetoQ::test_quantize_functions PASSED 2025-09-09T14:50:31.8075041Z test/prototype/test_paretoq.py::TestParetoQ::test_quantized_linear PASSED 2025-09-09T14:50:31.8075960Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8077161Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8078288Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8079410Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8080538Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8081662Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8082772Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8083887Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_0_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8085144Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8086276Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8087516Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8088637Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8089765Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8090898Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8092013Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8093142Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_1_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8094266Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8095398Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:50:31.8096528Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:50:31.8097642Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4076815Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:52:03.4078023Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4079164Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:52:03.4080290Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_2_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4081411Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:52:03.4082545Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4083698Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:52:03.4084818Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_False_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4085949Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_False_per_group_quantizer_False PASSED 2025-09-09T14:52:03.4087065Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_False_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4088184Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_True_per_group_quantizer_False PASSED 2025-09-09T14:52:03.4089642Z test/prototype/test_parq.py::TestPARQuantization::test_parq_train_loop_b_4_unif_quant_True_hard_prox_True_per_group_quantizer_True PASSED 2025-09-09T14:52:03.4090587Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_int4_weight_only_e2e PASSED 2025-09-09T14:52:03.4091608Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_int4_weight_only_group_size_256 PASSED 2025-09-09T14:52:03.4092480Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_int4_weight_only_group_size_32 PASSED 2025-09-09T14:52:03.4093353Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_2_group_size_32 PASSED 2025-09-09T14:52:03.4094246Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_2_group_size_512 PASSED 2025-09-09T14:52:03.4095132Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_3_group_size_32 PASSED 2025-09-09T14:52:03.4096020Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_3_group_size_512 PASSED 2025-09-09T14:52:03.4096936Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_4_group_size_32 PASSED 2025-09-09T14:52:03.4097835Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_4_group_size_512 PASSED 2025-09-09T14:52:03.4098726Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_8_group_size_32 PASSED 2025-09-09T14:52:03.4099612Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_b_8_group_size_512 PASSED 2025-09-09T14:52:03.4100462Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_2 PASSED 2025-09-09T14:52:03.4101287Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_3 PASSED 2025-09-09T14:52:03.4102098Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_4 PASSED 2025-09-09T14:52:03.4102924Z test/prototype/test_parq.py::TestUnifTorchaoQuantizer::test_intx_weight_only_e2e_b_8 PASSED 2025-09-09T14:52:03.4103775Z test/prototype/test_parq.py::TestStretchedUnifTorchaoQuantizer::test_intx_weight_only PASSED 2025-09-09T14:52:03.4104653Z test/prototype/test_parq.py::TestStretchedUnifTorchaoQuantizer::test_intx_weight_only_e2e PASSED 2025-09-09T14:52:03.4105591Z test/prototype/test_parq.py::TestStretchedUnifTorchaoQuantizer::test_intx_weight_only_parq_equivalent PASSED 2025-09-09T14:52:03.4106726Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float16_group_size_128 PASSED 2025-09-09T14:52:03.4107991Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float16_group_size_32 PASSED 2025-09-09T14:52:03.4109250Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float32_group_size_128 PASSED 2025-09-09T14:52:03.4110503Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_2_float32_group_size_32 PASSED 2025-09-09T14:52:03.4111765Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float16_group_size_128 PASSED 2025-09-09T14:52:03.4113022Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float16_group_size_32 PASSED 2025-09-09T14:52:03.4114321Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float32_group_size_128 PASSED 2025-09-09T14:52:03.4115579Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_3_float32_group_size_32 PASSED 2025-09-09T14:52:03.4117015Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float16_group_size_128 PASSED 2025-09-09T14:52:03.4118274Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float16_group_size_32 PASSED 2025-09-09T14:52:03.4119616Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float32_group_size_128 PASSED 2025-09-09T14:52:03.4120869Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_4_float32_group_size_32 PASSED 2025-09-09T14:52:03.4122131Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float16_group_size_128 PASSED 2025-09-09T14:52:03.4123396Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float16_group_size_32 PASSED 2025-09-09T14:52:03.4124701Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float32_group_size_128 PASSED 2025-09-09T14:52:03.4125960Z test/prototype/test_parq.py::TestInt8DynamicActivationTorchaoQuantizer::test_int8_dynamic_activation_intx_e2e_b_8_float32_group_size_32 PASSED 2025-09-09T14:52:03.4127054Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_bitnet_training_compile_False SKIPPED 2025-09-09T14:52:03.4128012Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_bitnet_training_compile_True SKIPPED 2025-09-09T14:52:03.4129129Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config0_module_swap_False PASSED 2025-09-09T14:52:03.4130366Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config0_module_swap_True PASSED 2025-09-09T14:52:03.4131614Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config1_module_swap_False PASSED 2025-09-09T14:52:03.4132861Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config1_module_swap_True PASSED 2025-09-09T14:52:03.4134099Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config2_module_swap_False PASSED 2025-09-09T14:52:03.4135344Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config2_module_swap_True PASSED 2025-09-09T14:52:03.4136592Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config3_module_swap_False PASSED 2025-09-09T14:52:03.4137824Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_False_config3_module_swap_True PASSED 2025-09-09T14:52:03.4139063Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config0_module_swap_False PASSED 2025-09-09T14:52:03.4140291Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config0_module_swap_True PASSED 2025-09-09T14:52:03.4141529Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config1_module_swap_False PASSED 2025-09-09T14:56:25.1711156Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config1_module_swap_True PASSED 2025-09-09T14:56:25.1713536Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config2_module_swap_False PASSED 2025-09-09T14:56:25.1715347Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config2_module_swap_True PASSED 2025-09-09T14:56:25.1716928Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config3_module_swap_False PASSED 2025-09-09T14:56:25.1718182Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_mixed_precision_training_compile_True_config3_module_swap_True PASSED 2025-09-09T14:56:25.1719294Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_stochastic_rounding_device_cpu PASSED 2025-09-09T14:56:25.1720294Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_stochastic_rounding_device_cuda PASSED 2025-09-09T14:56:25.1721393Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_False_device_cpu PASSED 2025-09-09T14:56:25.1722629Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_False_device_cuda PASSED 2025-09-09T14:56:25.1723806Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_True_device_cpu PASSED 2025-09-09T14:56:25.1724966Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims0_bias_True_device_cuda PASSED 2025-09-09T14:56:25.1726137Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_False_device_cpu PASSED 2025-09-09T14:56:25.1727308Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_False_device_cuda PASSED 2025-09-09T14:56:25.1728470Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_True_device_cpu PASSED 2025-09-09T14:56:25.1729638Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims1_bias_True_device_cuda PASSED 2025-09-09T14:56:25.1730808Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_False_device_cpu PASSED 2025-09-09T14:56:25.1731972Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_False_device_cuda PASSED 2025-09-09T14:56:25.1733138Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_True_device_cpu PASSED 2025-09-09T14:56:25.1734300Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_compile_leading_dims2_bias_True_device_cuda PASSED 2025-09-09T14:56:25.1735484Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_False_device_cpu PASSED 2025-09-09T14:56:25.1736695Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_False_device_cuda PASSED 2025-09-09T14:56:25.1737898Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_True_device_cpu PASSED 2025-09-09T14:56:25.1739100Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims0_bias_True_device_cuda PASSED 2025-09-09T14:56:25.1740325Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_False_device_cpu PASSED 2025-09-09T14:56:25.1741524Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_False_device_cuda PASSED 2025-09-09T14:56:25.1742828Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_True_device_cpu PASSED 2025-09-09T14:56:25.1744026Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims1_bias_True_device_cuda PASSED 2025-09-09T14:56:25.1745313Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_False_device_cpu PASSED 2025-09-09T14:56:25.1746519Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_False_device_cuda PASSED 2025-09-09T14:56:25.1747719Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_True_device_cpu PASSED 2025-09-09T14:56:25.1748908Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_correctness_leading_dims2_bias_True_device_cuda PASSED 2025-09-09T14:56:25.1750069Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_False_device_cpu PASSED 2025-09-09T14:56:25.1751185Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_False_device_cuda PASSED 2025-09-09T14:56:25.1752281Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_True_device_cpu PASSED 2025-09-09T14:56:25.1753381Z test/prototype/test_quantized_training.py::TestQuantizedTraining::test_int8_weight_only_training_compile_True_device_cuda PASSED 2025-09-09T14:56:25.1754580Z test/prototype/test_quantized_training.py::TestFSDP2::test_fsdp2_correctness I0909 14:55:16.542094 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 57982 2025-09-09T14:56:25.1755727Z I0909 14:55:16.603247 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 57983 2025-09-09T14:56:25.1756358Z dist init r=0, world=2 2025-09-09T14:56:25.1756578Z dist init r=1, world=2 2025-09-09T14:56:25.1766085Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1767012Z warnings.warn( # warn only once 2025-09-09T14:56:25.1767917Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1768828Z warnings.warn( # warn only once 2025-09-09T14:56:25.1769724Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1770641Z warnings.warn( # warn only once 2025-09-09T14:56:25.1771529Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1772442Z warnings.warn( # warn only once 2025-09-09T14:56:25.1773326Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1774235Z warnings.warn( # warn only once 2025-09-09T14:56:25.1775134Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1776041Z warnings.warn( # warn only once 2025-09-09T14:56:25.1777168Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1778077Z warnings.warn( # warn only once 2025-09-09T14:56:25.1779090Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1780003Z warnings.warn( # warn only once 2025-09-09T14:56:25.1780893Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1781800Z warnings.warn( # warn only once 2025-09-09T14:56:25.1782687Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T14:56:25.1783604Z warnings.warn( # warn only once 2025-09-09T14:56:25.1783948Z PASSED 2025-09-09T14:56:30.6112750Z test/prototype/test_quantized_training.py::TestFSDP2::test_precompute_bitnet_scale I0909 14:56:25.168992 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 58837 2025-09-09T14:56:30.6113955Z I0909 14:56:25.229513 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 58838 2025-09-09T14:56:30.6114506Z dist init r=0, world=2 2025-09-09T14:56:30.6114739Z dist init r=1, world=2 2025-09-09T14:56:30.6115187Z PASSED 2025-09-09T14:56:30.6115665Z test/prototype/test_scheduler.py::TestScheduler::test_constructor PASSED 2025-09-09T14:56:30.6116457Z test/prototype/test_scheduler.py::TestScheduler::test_lambda_scheduler PASSED 2025-09-09T14:56:30.6117161Z test/prototype/test_scheduler.py::TestScheduler::test_order_of_steps PASSED 2025-09-09T14:56:30.6117807Z test/prototype/test_scheduler.py::TestScheduler::test_step PASSED 2025-09-09T14:56:30.6118458Z test/prototype/test_scheduler.py::TestCubicScheduler::test_constructor PASSED 2025-09-09T14:56:30.6119141Z test/prototype/test_scheduler.py::TestCubicScheduler::test_step PASSED 2025-09-09T14:56:30.6120018Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6121098Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6122170Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6123244Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6124330Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6125398Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6126476Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6127542Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6128594Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6129656Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6131168Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6132379Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha0_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6133464Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6134533Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6135614Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6136697Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6137783Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6138867Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6139940Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6141006Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6142073Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6143139Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6144227Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6145295Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_5_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6146376Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6147462Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6148536Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6149620Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6150726Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6151808Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6152897Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6153968Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6155041Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6156205Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6159047Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6160116Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_save_load_recipe_alpha_0_75_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6161348Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6162537Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6163980Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6165191Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6166393Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6167599Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6168789Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6169965Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6171150Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6172339Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6173530Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6174721Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha0_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6175914Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6177122Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6458474Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6461517Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6464733Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6466545Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6467748Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6468957Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6470159Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6471537Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6472861Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6474069Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_5_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6475271Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6476633Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6477850Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6479059Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6480287Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6481493Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6482702Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6483902Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6485106Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6486322Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6487535Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6488734Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_False_alpha_0_75_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6489932Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6491129Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6492306Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6493507Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6494703Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6495897Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6497089Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6498358Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6499651Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6500847Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6502033Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6503221Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha0_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6504418Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6505616Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6506866Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6508064Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6509277Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6510481Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6511677Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6512874Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6514072Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6515257Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6516535Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6517725Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_5_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6518923Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cpu_bfloat16 SKIPPED 2025-09-09T14:56:30.6520138Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cpu_float16 SKIPPED 2025-09-09T14:56:30.6521331Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cpu_float32 SKIPPED 2025-09-09T14:56:30.6522532Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cuda_bfloat16 SKIPPED 2025-09-09T14:56:30.6523740Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cuda_float16 SKIPPED 2025-09-09T14:56:30.6525041Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_dynamic_device_cuda_float32 SKIPPED 2025-09-09T14:56:30.6526247Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cpu_bfloat16 SKIPPED 2025-09-09T14:58:12.1314358Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cpu_float16 SKIPPED 2025-09-09T14:58:12.1316626Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cpu_float32 SKIPPED 2025-09-09T14:58:12.1317892Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cuda_bfloat16 SKIPPED 2025-09-09T14:58:12.1319092Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cuda_float16 SKIPPED 2025-09-09T14:58:12.1320316Z test/prototype/test_smoothquant.py::TestSmoothQuant::test_smoothquant_accuracy_bias_True_alpha_0_75_quant_mode_static_device_cuda_float32 SKIPPED 2025-09-09T14:58:12.1321252Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_constructor PASSED 2025-09-09T14:58:12.1321957Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_convert PASSED 2025-09-09T14:58:12.1322652Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash PASSED 2025-09-09T14:58:12.1323420Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash_with_params1 PASSED 2025-09-09T14:58:12.1324244Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash_with_params2 PASSED 2025-09-09T14:58:12.1325060Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_mask_squash_with_params3 PASSED 2025-09-09T14:58:12.1325827Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_prepare_config PASSED 2025-09-09T14:58:12.1326544Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_state_dict PASSED 2025-09-09T14:58:12.1327219Z test/prototype/test_sparsifier.py::TestBaseSparsifier::test_step PASSED 2025-09-09T14:58:12.1327949Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_constructor PASSED 2025-09-09T14:58:12.1328730Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_mask_squash PASSED 2025-09-09T14:58:12.1329480Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_prepare PASSED 2025-09-09T14:58:12.1330263Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_sparsity_levels PASSED 2025-09-09T14:58:12.1331013Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_step PASSED 2025-09-09T14:58:12.1331754Z test/prototype/test_sparsifier.py::TestWeightNormSparsifier::test_step_2_of_4 PASSED 2025-09-09T14:58:12.1332549Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_constructor PASSED 2025-09-09T14:58:12.1333368Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_mask_squash PASSED 2025-09-09T14:58:12.1334172Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_prepare PASSED 2025-09-09T14:58:12.1334987Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_sparsity_levels PASSED 2025-09-09T14:58:12.1335784Z test/prototype/test_sparsifier.py::TestNearlyDiagonalSparsifier::test_step PASSED 2025-09-09T14:58:12.1336569Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_fqn_to_module PASSED 2025-09-09T14:58:12.1337420Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_fqn_to_module_fail PASSED 2025-09-09T14:58:12.1338318Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_fqn_to_module_for_tensors PASSED 2025-09-09T14:58:12.1339508Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_get_arg_info_from_tensor_fqn PASSED 2025-09-09T14:58:12.1340471Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_get_arg_info_from_tensor_fqn_fail PASSED 2025-09-09T14:58:12.1341451Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_module_to_fqn PASSED 2025-09-09T14:58:12.1342307Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_module_to_fqn_fail PASSED 2025-09-09T14:58:12.1343184Z test/prototype/test_sparsity_utils.py::TestSparsityUtilFunctions::test_module_to_fqn_root PASSED 2025-09-09T14:58:12.1343941Z test/prototype/test_spinquant.py::test_spinquant_no_quantization[cpu] PASSED 2025-09-09T14:58:12.1344624Z test/prototype/test_spinquant.py::test_spinquant_no_quantization[cuda] PASSED 2025-09-09T14:58:12.1345452Z test/prototype/test_structured_sparsifier.py::TestSaliencyPruner::test_lstm_saliency_pruner_update_mask PASSED 2025-09-09T14:58:12.1346421Z test/prototype/test_structured_sparsifier.py::TestSaliencyPruner::test_saliency_pruner_update_mask PASSED 2025-09-09T14:58:12.1347356Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_complex_conv2d PASSED 2025-09-09T14:58:12.1348282Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_constructor PASSED 2025-09-09T14:58:12.1349209Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prepare_conv2d PASSED 2025-09-09T14:58:12.1350143Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prepare_linear PASSED 2025-09-09T14:58:12.1351154Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_activation_conv2d PASSED 2025-09-09T14:58:12.1352191Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_bias_conv2d PASSED 2025-09-09T14:58:12.1353184Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_conv2d PASSED 2025-09-09T14:58:12.1354195Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_padding_conv2d PASSED 2025-09-09T14:58:12.1355228Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_conv2d_pool_conv2d PASSED 2025-09-09T14:58:12.1356356Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_linear_activation_linear PASSED 2025-09-09T14:58:12.1357887Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_linear_bias_linear PASSED 2025-09-09T14:58:12.1359327Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_linear_linear PASSED 2025-09-09T14:58:12.1360411Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_layernorm_linear_multiple_layer PASSED 2025-09-09T14:58:12.1361555Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_layernorm_linear_single_layer PASSED 2025-09-09T14:58:12.1362655Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_linear_multiple_layer PASSED 2025-09-09T14:58:12.1363952Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_prune_lstm_linear_single_layer PASSED 2025-09-09T14:58:12.1364928Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_step_conv2d PASSED 2025-09-09T14:58:12.1365840Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_step_linear PASSED 2025-09-09T14:58:12.1366699Z test/prototype/test_structured_sparsifier.py::TestFPGMPruner::test_compute_distance PASSED 2025-09-09T14:58:12.1367485Z test/prototype/test_structured_sparsifier.py::TestFPGMPruner::test_update_mask PASSED 2025-09-09T14:58:12.1368556Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_attention_block SKIPPED 2025-09-09T14:58:12.1369633Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_conv2d SKIPPED 2025-09-09T14:58:12.1370608Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_conv2d_binary SKIPPED 2025-09-09T14:58:12.1371605Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_conv2d_binary2 SKIPPED 2025-09-09T14:58:12.1372626Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_dynamic_quant_linear SKIPPED 2025-09-09T14:58:12.1373673Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_filter_linear_recipe SKIPPED 2025-09-09T14:58:12.1374664Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear SKIPPED 2025-09-09T14:58:12.1375616Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary SKIPPED 2025-09-09T14:58:12.1376638Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary_dynamic SKIPPED 2025-09-09T14:58:12.1377693Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary_dynamic_qat SKIPPED 2025-09-09T14:58:12.1378739Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_linear_unary_qat SKIPPED 2025-09-09T14:58:12.1379724Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_conv2d SKIPPED 2025-09-09T14:58:12.1380715Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_conv2d_binary SKIPPED 2025-09-09T14:58:12.1381755Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_conv2d_binary2 SKIPPED 2025-09-09T14:58:12.1382799Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_qat_dynamic_quant_linear SKIPPED 2025-09-09T14:59:13.5464299Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_and_module_type_case1 SKIPPED 2025-09-09T14:59:13.5465498Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_and_module_type_case2 SKIPPED 2025-09-09T14:59:13.5466740Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_and_module_type_with_mixed_configs SKIPPED 2025-09-09T14:59:13.5467930Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_qconfig SKIPPED 2025-09-09T14:59:13.5469074Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_qconfig_for_dynamic_quant SKIPPED 2025-09-09T14:59:13.5470268Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_qconfig_with_underscores SKIPPED 2025-09-09T14:59:13.5471430Z test/quantization/pt2e/test_arm_inductor_quantizer.py::TestQuantizePT2EArmInductor::test_set_module_name_with_mixed_configs SKIPPED 2025-09-09T14:59:13.5472464Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_avgpool_use_different_qconfig PASSED 2025-09-09T14:59:13.5473394Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_no_add_quant_duplicate_dq PASSED 2025-09-09T14:59:13.5474296Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_no_need_for_duplicate_dq PASSED 2025-09-09T14:59:13.5475185Z test/quantization/pt2e/test_duplicate_dq.py::TestDuplicateDQPass::test_simple_duplicate_dq PASSED 2025-09-09T14:59:13.5476398Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_conv_relu PASSED 2025-09-09T14:59:13.5477631Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_relu W0909 14:58:19.112326 915 site-packages/torch/fx/experimental/symbolic_shapes.py:3144] Failed to reduce inequalities: 1/2 2025-09-09T14:59:13.5478513Z PASSED 2025-09-09T14:59:13.5479365Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_customized_equivalet_types_dict W0909 14:58:19.280255 915 site-packages/torch/fx/experimental/symbolic_shapes.py:3144] Failed to reduce inequalities: 1/2 2025-09-09T14:59:13.5480308Z PASSED 2025-09-09T14:59:13.5480910Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_for_dq SKIPPED 2025-09-09T14:59:13.5481963Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_for_dq_no_static_q PASSED 2025-09-09T14:59:13.5482974Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_for_two_dq PASSED 2025-09-09T14:59:13.5483998Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_metadata_porting_with_no_quant_inbetween PASSED 2025-09-09T14:59:13.5484987Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_no_metadata_porting PASSED 2025-09-09T14:59:13.5485991Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_no_metadata_porting_through_unknown_ops PASSED 2025-09-09T14:59:13.5486982Z test/quantization/pt2e/test_metadata_porting.py::TestMetaDataPorting::test_simple_metadata_porting PASSED 2025-09-09T14:59:13.5487966Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_added_node_gets_unique_id PASSED 2025-09-09T14:59:13.5488923Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_control_flow SKIPPED 2025-09-09T14:59:13.5489868Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_copy_preserve_handle PASSED 2025-09-09T14:59:13.5490858Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_deepcopy_preserve_handle PASSED 2025-09-09T14:59:13.5491905Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_prepare_for_propagation_comparison PASSED 2025-09-09T14:59:13.5492947Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_re_export_preserve_handle PASSED 2025-09-09T14:59:13.5494027Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_run_decompositions_map_handle_to_new_nodes PASSED 2025-09-09T14:59:13.5495126Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_run_decompositions_same_handle_id PASSED 2025-09-09T14:59:13.5496082Z test/quantization/pt2e/test_numeric_debugger.py::TestNumericDebuggerInfra::test_simple PASSED 2025-09-09T14:59:13.5496981Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_exported_model_train_eval PASSED 2025-09-09T14:59:13.5497966Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_exported_model_train_eval_idempotent PASSED 2025-09-09T14:59:13.5498911Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_implicit_sharing PASSED 2025-09-09T14:59:13.5499762Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_chunked_bn_fusion PASSED 2025-09-09T14:59:13.5501615Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_composable_quantizer_linear_conv [W909 14:59:13.849323469 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5504340Z [W909 14:59:13.849363991 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5506490Z [W909 14:59:13.849403633 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5508622Z [W909 14:59:13.849421904 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5510770Z [W909 14:59:13.849446486 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5512910Z [W909 14:59:13.849473237 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5515045Z [W909 14:59:13.849500589 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5517257Z [W909 14:59:13.849518760 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5519401Z [W909 14:59:13.849608675 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5521540Z [W909 14:59:13.849633836 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T14:59:13.5523689Z [W909 14:59:13.849654257 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T15:00:27.4888749Z [W909 14:59:13.849668418 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T15:00:27.4891628Z [W909 14:59:13.849686009 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T15:00:27.4894990Z [W909 14:59:13.849695859 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T15:00:27.4897748Z [W909 14:59:13.849802885 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T15:00:27.4900548Z [W909 14:59:13.849827907 PyInterpreter.cpp:263] Warning: Deallocating Tensor that still has live PyObject references. This probably happened because you took out a weak reference to Tensor and didn't call _fix_weakref() after dereferencing it. Subsequent accesses to this tensor via the PyObject will now fail. (function decref) 2025-09-09T15:00:27.4902331Z PASSED 2025-09-09T15:00:27.4903083Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_composable_quantizer_throw PASSED 2025-09-09T15:00:27.4904344Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_composable_quantizer_transform_for_annotation PASSED 2025-09-09T15:00:27.4905621Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_constant_prop_preserve_metadata PASSED 2025-09-09T15:00:27.4906725Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv3d_bn_relu PASSED 2025-09-09T15:00:27.4907771Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv_padding_bn_relu PASSED 2025-09-09T15:00:27.4908891Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv_transpose3d_bn_relu PASSED 2025-09-09T15:00:27.4910007Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_conv_transpose_bn_relu PASSED 2025-09-09T15:00:27.4911067Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_derived_qspec PASSED 2025-09-09T15:00:27.4912146Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_derived_qspec_per_channel PASSED 2025-09-09T15:00:27.4913239Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_disallow_eval_train PASSED 2025-09-09T15:00:27.4914341Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_dont_fold_other_constant PASSED 2025-09-09T15:00:27.4915518Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_embedding_conv_linear_quantization PASSED 2025-09-09T15:00:27.4916816Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_embedding_quantizer PASSED 2025-09-09T15:00:27.4917949Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fixed_qparams_qspec_observer_dedup PASSED 2025-09-09T15:00:27.4919093Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fixed_qparams_qspec_ptq PASSED 2025-09-09T15:00:27.4920190Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fixed_qparams_qspec_qat PASSED 2025-09-09T15:00:27.4929669Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_all_ops_before_quantize PASSED 2025-09-09T15:00:27.4930772Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_quantize PASSED 2025-09-09T15:00:27.4931845Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_fold_quantize_per_channel PASSED 2025-09-09T15:00:27.4933110Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_groupwise_per_channel_quant PASSED 2025-09-09T15:00:27.4934226Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_input_edge_sanity_check PASSED 2025-09-09T15:00:27.4935408Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_max_pool2d_quantizer PASSED 2025-09-09T15:00:27.4936486Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_model_is_exported PASSED 2025-09-09T15:00:27.4937609Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_bn_device_cpu PASSED 2025-09-09T15:00:27.4938812Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_bn_device_cuda PASSED 2025-09-09T15:00:27.4940033Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_dropout PASSED 2025-09-09T15:00:27.4941228Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_move_exported_model_dropout_inplace PASSED 2025-09-09T15:00:27.4942473Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_multi_users_without_output_observer PASSED 2025-09-09T15:00:27.4943614Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_observer_callback PASSED 2025-09-09T15:00:27.4944721Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_prepare_obs_or_fq_callback PASSED 2025-09-09T15:00:27.4945856Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_preserve_nn_module_stack PASSED 2025-09-09T15:00:27.4947069Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_bfloat16_float8_e4m3fn PASSED 2025-09-09T15:00:27.4948355Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_bfloat16_float8_e5m2 PASSED 2025-09-09T15:00:27.4949658Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_bfloat16_int16 PASSED 2025-09-09T15:00:27.4950848Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_float32_float8_e4m3fn PASSED 2025-09-09T15:00:27.4951848Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_float32_float8_e5m2 PASSED 2025-09-09T15:00:27.4952939Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantization_dtype_float32_int16 PASSED 2025-09-09T15:00:27.4954016Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_quantize_in_place_ops input_act1 is a node 2025-09-09T15:00:27.4954735Z PASSED 2025-09-09T15:00:27.4955354Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_reentrant PASSED 2025-09-09T15:00:27.4956962Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_save_load W0909 15:00:21.289155 915 site-packages/torch/export/pt2_archive/_package.py:406] Expect archive file to be a file ending in .pt2, or is a buffer. Instead got {/tmp/tmp_rri7hb3} 2025-09-09T15:00:27.4958808Z W0909 15:00:21.300090 915 site-packages/torch/export/pt2_archive/_package.py:583] Unable to load package. f must be a buffer or a file ending in .pt2. Instead got {/tmp/tmp_rri7hb3} 2025-09-09T15:00:27.4959756Z PASSED 2025-09-09T15:00:27.4960385Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_shared_qspec PASSED 2025-09-09T15:00:27.4961460Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_shared_qspec_transitivity PASSED 2025-09-09T15:00:27.4962640Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_shared_qspec_transitivity_case_2 PASSED 2025-09-09T15:00:27.4964125Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_simple_quantizer PASSED 2025-09-09T15:00:27.4965117Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_speed PASSED 2025-09-09T15:00:27.4966316Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_transform_for_annotation PASSED 2025-09-09T15:00:27.4967474Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_wo_annotate_conv_output_quantizer PASSED 2025-09-09T15:00:27.4968837Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2EAffineQuantization::test_channel_group_quantization prepared model: GraphModule( 2025-09-09T15:00:27.4969689Z (linear): Module() 2025-09-09T15:00:27.4970074Z (activation_post_process_1): AffineQuantizedMinMaxObserver() 2025-09-09T15:00:27.4970622Z (activation_post_process_0): AffineQuantizedMinMaxObserver() 2025-09-09T15:00:27.4971049Z ) 2025-09-09T15:00:27.4971170Z 2025-09-09T15:00:27.4971175Z 2025-09-09T15:00:27.4971180Z 2025-09-09T15:00:27.4971285Z def forward(self, x): 2025-09-09T15:00:27.4971626Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:00:27.4972043Z linear_weight = self.linear.weight 2025-09-09T15:00:27.4972650Z activation_post_process_1 = self.activation_post_process_1(linear_weight); linear_weight = None 2025-09-09T15:00:27.4973263Z linear_bias = self.linear.bias 2025-09-09T15:00:27.4973725Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:00:27.4974836Z linear = torch.ops.aten.linear.default(activation_post_process_0, activation_post_process_1, linear_bias); activation_post_process_0 = activation_post_process_1 = linear_bias = None 2025-09-09T15:00:27.4975866Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T15:00:27.4976260Z 2025-09-09T15:01:13.2725831Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:13.2726332Z quantized model GraphModule( 2025-09-09T15:01:13.2726655Z (linear): Module() 2025-09-09T15:01:13.2726919Z ) 2025-09-09T15:01:13.2727059Z 2025-09-09T15:01:13.2727064Z 2025-09-09T15:01:13.2727070Z 2025-09-09T15:01:13.2727181Z def forward(self, x): 2025-09-09T15:01:13.2727563Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:13.2727969Z _scale0 = self._scale0 2025-09-09T15:01:13.2728287Z _zero_point0 = self._zero_point0 2025-09-09T15:01:13.2728641Z quantize_affine = self._frozen_param0 2025-09-09T15:01:13.2729736Z dequantize_affine = torch.ops.torchao.dequantize_affine(quantize_affine, (1, 128), _scale0, _zero_point0, torch.uint8, 0, 255, output_dtype = torch.float32); quantize_affine = _scale0 = _zero_point0 = None 2025-09-09T15:01:13.2730790Z linear_bias = self.linear.bias 2025-09-09T15:01:13.2731117Z _scale1 = self._scale1 2025-09-09T15:01:13.2731453Z _zero_point1 = self._zero_point1 2025-09-09T15:01:13.2732116Z quantize_affine_1 = torch.ops.torchao.quantize_affine(x, (1, 128), _scale1, _zero_point1, torch.uint8, 0, 255); x = None 2025-09-09T15:01:13.2733489Z dequantize_affine_1 = torch.ops.torchao.dequantize_affine(quantize_affine_1, (1, 128), _scale1, _zero_point1, torch.uint8, 0, 255, output_dtype = torch.float32); quantize_affine_1 = _scale1 = _zero_point1 = None 2025-09-09T15:01:13.2735052Z linear = torch.ops.aten.linear.default(dequantize_affine_1, dequantize_affine, linear_bias); dequantize_affine_1 = dequantize_affine = linear_bias = None 2025-09-09T15:01:13.2735989Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T15:01:13.2736395Z 2025-09-09T15:01:13.2736725Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:13.2737392Z PASSED 2025-09-09T15:01:13.2738339Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2EAffineQuantization::test_dynamic_affine_act_per_channel_weights PASSED 2025-09-09T15:01:13.2739836Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2EAffineQuantization::test_dynamic_per_tok_act_per_group_weights prepared model: GraphModule( 2025-09-09T15:01:13.2740742Z (linear): Module() 2025-09-09T15:01:13.2741120Z (activation_post_process_1): AffineQuantizedMinMaxObserver() 2025-09-09T15:01:13.2742101Z (activation_post_process_0): AffineQuantizedPlaceholderObserver() 2025-09-09T15:01:13.2742544Z ) 2025-09-09T15:01:13.2742669Z 2025-09-09T15:01:13.2742675Z 2025-09-09T15:01:13.2742680Z 2025-09-09T15:01:13.2742780Z def forward(self, x): 2025-09-09T15:01:13.2743298Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:13.2743718Z linear_weight = self.linear.weight 2025-09-09T15:01:13.2744321Z activation_post_process_1 = self.activation_post_process_1(linear_weight); linear_weight = None 2025-09-09T15:01:13.2744930Z linear_bias = self.linear.bias 2025-09-09T15:01:13.2745403Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:01:13.2746522Z linear = torch.ops.aten.linear.default(activation_post_process_0, activation_post_process_1, linear_bias); activation_post_process_0 = activation_post_process_1 = linear_bias = None 2025-09-09T15:01:13.2747607Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T15:01:13.2748018Z 2025-09-09T15:01:13.2748351Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:13.2748832Z quantized model GraphModule( 2025-09-09T15:01:13.2749158Z (linear): Module() 2025-09-09T15:01:13.2749399Z ) 2025-09-09T15:01:13.2749515Z 2025-09-09T15:01:13.2749519Z 2025-09-09T15:01:13.2749531Z 2025-09-09T15:01:13.2749638Z def forward(self, x): 2025-09-09T15:01:13.2749964Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:13.2750358Z _scale0 = self._scale0 2025-09-09T15:01:13.2750645Z _zero_point0 = self._zero_point0 2025-09-09T15:01:13.2750991Z quantize_affine = self._frozen_param0 2025-09-09T15:01:13.2752017Z dequantize_affine = torch.ops.torchao.dequantize_affine(quantize_affine, (1, 128), _scale0, _zero_point0, torch.int8, -127, 127, output_dtype = torch.float32); quantize_affine = _scale0 = _zero_point0 = None 2025-09-09T15:01:13.2753027Z linear_bias = self.linear.bias 2025-09-09T15:01:13.2753724Z choose_qparams_affine = torch.ops.torchao.choose_qparams_affine(x, 'SYMMETRIC', (1, 128), torch.int8, -128, 127, None, None, None) 2025-09-09T15:01:13.2754444Z getitem = choose_qparams_affine[0] 2025-09-09T15:01:13.2754895Z getitem_1 = choose_qparams_affine[1]; choose_qparams_affine = None 2025-09-09T15:01:13.2755671Z quantize_affine_1 = torch.ops.torchao.quantize_affine(x, (1, 128), getitem, getitem_1, torch.int8, -128, 127); x = None 2025-09-09T15:01:13.2757073Z dequantize_affine_1 = torch.ops.torchao.dequantize_affine(quantize_affine_1, (1, 128), getitem, getitem_1, torch.int8, -128, 127, output_dtype = torch.float32); quantize_affine_1 = getitem = getitem_1 = None 2025-09-09T15:01:13.2758612Z linear = torch.ops.aten.linear.default(dequantize_affine_1, dequantize_affine, linear_bias); dequantize_affine_1 = dequantize_affine = linear_bias = None 2025-09-09T15:01:13.2759542Z return pytree.tree_unflatten((linear,), self._out_spec) 2025-09-09T15:01:13.2759952Z 2025-09-09T15:01:13.2760292Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:13.2760774Z PASSED 2025-09-09T15:01:13.2761605Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_fold_bn_erases_bn_node SKIPPED 2025-09-09T15:01:13.2762965Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_bias_derived_qspec SKIPPED 2025-09-09T15:01:13.2764263Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion SKIPPED 2025-09-09T15:01:13.2765278Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion_cuda SKIPPED 2025-09-09T15:01:13.2766329Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion_literal_args SKIPPED 2025-09-09T15:01:13.2767403Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_fusion_no_conv_bias SKIPPED 2025-09-09T15:01:13.2768643Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_per_channel_weight_bias SKIPPED 2025-09-09T15:01:13.2769823Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_relu_fusion SKIPPED 2025-09-09T15:01:13.2770875Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_relu_fusion_cuda SKIPPED 2025-09-09T15:01:13.2771948Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_bn_relu_fusion_no_conv_bias SKIPPED 2025-09-09T15:01:13.2772984Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_no_bias SKIPPED 2025-09-09T15:01:13.2773974Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_transpose_bn SKIPPED 2025-09-09T15:01:13.2775012Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_conv_transpose_bn_relu SKIPPED 2025-09-09T15:01:13.2776040Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_inplace_add_relu SKIPPED 2025-09-09T15:01:13.2777094Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_per_channel_weight_custom_dtype SKIPPED 2025-09-09T15:01:13.2778183Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_preserve_source_fn_stack SKIPPED 2025-09-09T15:01:13.2779272Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn_Base::test_qat_update_shared_qspec SKIPPED 2025-09-09T15:01:13.2780274Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_fold_bn_erases_bn_node PASSED 2025-09-09T15:01:13.2781296Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_bias_derived_qspec PASSED 2025-09-09T15:01:13.2782232Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion model pt2e: GraphModule( 2025-09-09T15:01:13.2782828Z (conv): Module() 2025-09-09T15:01:13.2783038Z (bn): Module() 2025-09-09T15:01:13.2783332Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:13.2784258Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:13.2785342Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:01:13.2785847Z ) 2025-09-09T15:01:13.2786119Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:13.2787100Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0017, 0.0019, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:01:13.2788376Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2148, -0.0992, -0.2048]), max_val=tensor([0.0771, 0.2459, 0.3011])) 2025-09-09T15:01:13.2789009Z ) 2025-09-09T15:01:13.2789285Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:13.2790205Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:13.2791289Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4139440059661865) 2025-09-09T15:01:13.2791923Z ) 2025-09-09T15:01:13.2792088Z ) 2025-09-09T15:01:13.2792189Z 2025-09-09T15:01:13.2792194Z 2025-09-09T15:01:13.2792198Z 2025-09-09T15:01:13.2792283Z def forward(self, x): 2025-09-09T15:01:13.2792561Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:13.2792902Z conv_weight = self.conv.weight 2025-09-09T15:01:13.2793253Z conv_bias = self.conv.bias 2025-09-09T15:01:13.2793512Z bn_weight = self.bn.weight 2025-09-09T15:01:13.2793766Z bn_bias = self.bn.bias 2025-09-09T15:01:13.2794019Z bn_running_mean = self.bn.running_mean 2025-09-09T15:01:13.2794319Z bn_running_var = self.bn.running_var 2025-09-09T15:01:13.2794644Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:01:13.2795086Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:01:32.3926099Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:01:32.3926831Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:01:32.3927369Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:01:32.3927887Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:01:32.3928444Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:01:32.3929100Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:01:32.3929788Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:01:32.3930580Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:01:32.3931811Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:01:32.3932950Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:01:32.3933628Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:01:32.3934342Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:01:32.3935037Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:01:32.3936132Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:01:32.3937289Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:01:32.3938025Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:01:32.3938515Z 2025-09-09T15:01:32.3938859Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:32.3939320Z model fx: GraphModule( 2025-09-09T15:01:32.3939735Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:32.3940899Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:32.3942349Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:01:32.3942990Z ) 2025-09-09T15:01:32.3943234Z (conv): ConvBn1d( 2025-09-09T15:01:32.3943527Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:01:32.3944028Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:01:32.3944618Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:32.3945807Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0017, 0.0019, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:01:32.3947739Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2148, -0.0992, -0.2048]), max_val=tensor([0.0771, 0.2459, 0.3011])) 2025-09-09T15:01:32.3948572Z ) 2025-09-09T15:01:32.3948785Z ) 2025-09-09T15:01:32.3949365Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:32.3950530Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:32.3951902Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4139440059661865) 2025-09-09T15:01:32.3952536Z ) 2025-09-09T15:01:32.3952749Z ) 2025-09-09T15:01:32.3952867Z 2025-09-09T15:01:32.3952872Z 2025-09-09T15:01:32.3952877Z 2025-09-09T15:01:32.3952994Z def forward(self, x): 2025-09-09T15:01:32.3953413Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:01:32.3954067Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:01:32.3954747Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:01:32.3955286Z return activation_post_process_1 2025-09-09T15:01:32.3955598Z 2025-09-09T15:01:32.3955940Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:32.3956461Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:01:32.3956739Z [0., 0., 0.], 2025-09-09T15:01:32.3957026Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:01:32.3957392Z converted model pt2e: GraphModule( 2025-09-09T15:01:32.3957713Z (conv): Module() 2025-09-09T15:01:32.3957957Z (bn): Module() 2025-09-09T15:01:32.3958191Z ) 2025-09-09T15:01:32.3958309Z 2025-09-09T15:01:32.3958314Z 2025-09-09T15:01:32.3958325Z 2025-09-09T15:01:32.3958427Z def forward(self, x): 2025-09-09T15:01:32.3958766Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:32.3959173Z conv_bias = self.conv.bias 2025-09-09T15:01:32.3959529Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:01:32.3960421Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:01:32.3961988Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:01:32.3963303Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:01:32.3964060Z _scale_0 = self._scale_0 2025-09-09T15:01:32.3964370Z _zero_point_0 = self._zero_point_0 2025-09-09T15:01:32.3964737Z quantize_per_channel = self._frozen_param0 2025-09-09T15:01:32.3965833Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:01:32.3967606Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:01:32.3968795Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011089790612459183, 0, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:01:32.3970061Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011089790612459183, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:01:32.3971065Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:01:32.3971662Z 2025-09-09T15:01:32.3971940Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:32.3972317Z onverted model fx: GraphModule( 2025-09-09T15:01:32.3972687Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:01:32.3973060Z ) 2025-09-09T15:01:32.3973276Z 2025-09-09T15:01:32.3973281Z 2025-09-09T15:01:32.3973285Z 2025-09-09T15:01:32.3973372Z def forward(self, x): 2025-09-09T15:01:32.3973984Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:01:32.3975203Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:01:32.3976196Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:01:32.3977047Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011089790612459183, 0, -128, 127, torch.int8); conv = None 2025-09-09T15:01:32.3978303Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011089790612459183, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:01:32.3979169Z return dequantize_per_tensor_default_1 2025-09-09T15:01:32.3979448Z 2025-09-09T15:01:32.3979722Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:32.3980094Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:01:32.3980319Z [0., 0., 0.], 2025-09-09T15:01:32.3980531Z [0., 0., 0.]]]) 2025-09-09T15:01:32.3980755Z model pt2e: GraphModule( 2025-09-09T15:01:32.3980984Z (conv): Module() 2025-09-09T15:01:32.3981181Z (bn): Module() 2025-09-09T15:01:32.3981483Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:32.3982414Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:32.3983497Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:01:32.3984006Z ) 2025-09-09T15:01:32.3984280Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:32.3985218Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:01:32.3986313Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.214811772108078, max_val=0.30109599232673645) 2025-09-09T15:01:32.3986821Z ) 2025-09-09T15:01:32.3987100Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:32.3988014Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:32.3989097Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4138529300689697) 2025-09-09T15:01:32.3989608Z ) 2025-09-09T15:01:32.3989774Z ) 2025-09-09T15:01:32.3989876Z 2025-09-09T15:01:32.3989886Z 2025-09-09T15:01:32.3989890Z 2025-09-09T15:01:32.3989974Z def forward(self, x): 2025-09-09T15:01:32.3990253Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:32.3990595Z conv_weight = self.conv.weight 2025-09-09T15:01:32.3990868Z conv_bias = self.conv.bias 2025-09-09T15:01:32.3991150Z bn_weight = self.bn.weight 2025-09-09T15:01:32.3991464Z bn_bias = self.bn.bias 2025-09-09T15:01:32.3991889Z bn_running_mean = self.bn.running_mean 2025-09-09T15:01:32.3992266Z bn_running_var = self.bn.running_var 2025-09-09T15:01:57.2777920Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:01:57.2778632Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:01:57.2779841Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:01:57.2780586Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:01:57.2781117Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:01:57.2781660Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:01:57.2782134Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:01:57.2782630Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:01:57.2783197Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:01:57.2783822Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:01:57.2784821Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:01:57.2785698Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:01:57.2786239Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:01:57.2786800Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:01:57.2787349Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:01:57.2788204Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:01:57.2789138Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:01:57.2789735Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:01:57.2790112Z 2025-09-09T15:01:57.2790396Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:57.2790752Z model fx: GraphModule( 2025-09-09T15:01:57.2791076Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:57.2792011Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:57.2793174Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:01:57.2793675Z ) 2025-09-09T15:01:57.2793861Z (conv): ConvBn1d( 2025-09-09T15:01:57.2794075Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:01:57.2794482Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:01:57.2794938Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:57.2795853Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:01:57.2797038Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.214811772108078, max_val=0.30109599232673645) 2025-09-09T15:01:57.2797600Z ) 2025-09-09T15:01:57.2797773Z ) 2025-09-09T15:01:57.2798043Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:57.2798970Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0111]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:57.2800241Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4139524698257446, max_val=1.4138529300689697) 2025-09-09T15:01:57.2800752Z ) 2025-09-09T15:01:57.2800918Z ) 2025-09-09T15:01:57.2801019Z 2025-09-09T15:01:57.2801106Z 2025-09-09T15:01:57.2801111Z 2025-09-09T15:01:57.2801200Z def forward(self, x): 2025-09-09T15:01:57.2801551Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:01:57.2802076Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:01:57.2802622Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:01:57.2803043Z return activation_post_process_1 2025-09-09T15:01:57.2803302Z 2025-09-09T15:01:57.2803581Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:57.2803955Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:01:57.2804192Z [0., 0., 0.], 2025-09-09T15:01:57.2804425Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:01:57.2804730Z converted model pt2e: GraphModule( 2025-09-09T15:01:57.2804991Z (conv): Module() 2025-09-09T15:01:57.2805198Z (bn): Module() 2025-09-09T15:01:57.2805387Z ) 2025-09-09T15:01:57.2805497Z 2025-09-09T15:01:57.2805502Z 2025-09-09T15:01:57.2805506Z 2025-09-09T15:01:57.2805591Z def forward(self, x): 2025-09-09T15:01:57.2805877Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:01:57.2806202Z conv_bias = self.conv.bias 2025-09-09T15:01:57.2806506Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:01:57.2807201Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:01:57.2808479Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:01:57.2809520Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:01:57.2810005Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:01:57.2810795Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002370834583416581, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:01:57.2812304Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:01:57.2813491Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.01108943298459053, 0, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:01:57.2814756Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01108943298459053, 0, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:01:57.2815939Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:01:57.2816349Z 2025-09-09T15:01:57.2816626Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:57.2817001Z onverted model fx: GraphModule( 2025-09-09T15:01:57.2817369Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:01:57.2817730Z ) 2025-09-09T15:01:57.2817823Z 2025-09-09T15:01:57.2817827Z 2025-09-09T15:01:57.2817831Z 2025-09-09T15:01:57.2817921Z def forward(self, x): 2025-09-09T15:01:57.2818572Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:01:57.2819785Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:01:57.2820900Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:01:57.2821814Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01108943298459053, 0, -128, 127, torch.int8); conv = None 2025-09-09T15:01:57.2823065Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01108943298459053, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:01:57.2823931Z return dequantize_per_tensor_default_1 2025-09-09T15:01:57.2824197Z 2025-09-09T15:01:57.2824471Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:01:57.2824828Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:01:57.2825056Z [0., 0., 0.], 2025-09-09T15:01:57.2825263Z [0., 0., 0.]]]) 2025-09-09T15:01:57.2825709Z PASSED 2025-09-09T15:01:57.2826290Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion_cuda model pt2e: GraphModule( 2025-09-09T15:01:57.2826912Z (conv): Module() 2025-09-09T15:01:57.2827110Z (bn): Module() 2025-09-09T15:01:57.2827419Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:57.2828526Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:01:57.2829774Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:01:57.2830282Z ) 2025-09-09T15:01:57.2830551Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:57.2831712Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:01:57.2833246Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T15:01:57.2833965Z ) 2025-09-09T15:01:57.2834237Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:01:57.2835348Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:16.4928888Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3958139419555664, max_val=1.4123148918151855) 2025-09-09T15:02:16.4929656Z ) 2025-09-09T15:02:16.4929885Z ) 2025-09-09T15:02:16.4930008Z 2025-09-09T15:02:16.4930014Z 2025-09-09T15:02:16.4930019Z 2025-09-09T15:02:16.4930166Z def forward(self, x): 2025-09-09T15:02:16.4930535Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:02:16.4931010Z conv_weight = self.conv.weight 2025-09-09T15:02:16.4931384Z conv_bias = self.conv.bias 2025-09-09T15:02:16.4931736Z bn_weight = self.bn.weight 2025-09-09T15:02:16.4932070Z bn_bias = self.bn.bias 2025-09-09T15:02:16.4932417Z bn_running_mean = self.bn.running_mean 2025-09-09T15:02:16.4932715Z bn_running_var = self.bn.running_var 2025-09-09T15:02:16.4933054Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:02:16.4933581Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:02:16.4934604Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:02:16.4935132Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:02:16.4935518Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:02:16.4936083Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:02:16.4936519Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:02:16.4937022Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:02:16.4937587Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:02:16.4938192Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:02:16.4939155Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:02:16.4940082Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:02:16.4940614Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:02:16.4941189Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:02:16.4941728Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:02:16.4942593Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:02:16.4943504Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:02:16.4944106Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:02:16.4944494Z 2025-09-09T15:02:16.4944774Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:16.4945143Z model fx: GraphModule( 2025-09-09T15:02:16.4945466Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:16.4946580Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:16.4947852Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:02:16.4948352Z ) 2025-09-09T15:02:16.4948534Z (conv): ConvBn1d( 2025-09-09T15:02:16.4948748Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:02:16.4949146Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:02:16.4949608Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:16.4950786Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:02:16.4952562Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T15:02:16.4953290Z ) 2025-09-09T15:02:16.4953463Z ) 2025-09-09T15:02:16.4953732Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:16.4954837Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:16.4956565Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3958139419555664, max_val=1.4123148918151855) 2025-09-09T15:02:16.4957069Z ) 2025-09-09T15:02:16.4957237Z ) 2025-09-09T15:02:16.4957332Z 2025-09-09T15:02:16.4957433Z 2025-09-09T15:02:16.4957438Z 2025-09-09T15:02:16.4957526Z def forward(self, x): 2025-09-09T15:02:16.4957875Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:02:16.4958411Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:02:16.4958952Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:02:16.4959387Z return activation_post_process_1 2025-09-09T15:02:16.4959636Z 2025-09-09T15:02:16.4959913Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:16.4960296Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:02:16.4960570Z [0., 0., 0.], 2025-09-09T15:02:16.4960829Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T15:02:16.4961159Z converted model pt2e: GraphModule( 2025-09-09T15:02:16.4961421Z (conv): Module() 2025-09-09T15:02:16.4961616Z (bn): Module() 2025-09-09T15:02:16.4961807Z ) 2025-09-09T15:02:16.4961911Z 2025-09-09T15:02:16.4961916Z 2025-09-09T15:02:16.4961920Z 2025-09-09T15:02:16.4962002Z def forward(self, x): 2025-09-09T15:02:16.4962281Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:02:16.4962607Z conv_bias = self.conv.bias 2025-09-09T15:02:16.4962902Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:02:16.4963599Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:02:16.4965083Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:02:16.4966118Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:02:16.4966577Z _scale_0 = self._scale_0 2025-09-09T15:02:16.4966835Z _zero_point_0 = self._zero_point_0 2025-09-09T15:02:16.4967139Z quantize_per_channel = self._frozen_param0 2025-09-09T15:02:16.4968008Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:02:16.4969345Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:02:16.4970531Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011012271046638489, -1, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:02:16.4971817Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011012271046638489, -1, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:02:16.4972826Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:02:16.4973228Z 2025-09-09T15:02:16.4973510Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:16.4973877Z onverted model fx: GraphModule( 2025-09-09T15:02:16.4974247Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:02:16.4974616Z ) 2025-09-09T15:02:16.4974712Z 2025-09-09T15:02:16.4974716Z 2025-09-09T15:02:16.4974738Z 2025-09-09T15:02:16.4974821Z def forward(self, x): 2025-09-09T15:02:16.4975424Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:02:16.4976796Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:02:16.4977888Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:02:16.4978735Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011012271046638489, -1, -128, 127, torch.int8); conv = None 2025-09-09T15:02:16.4979981Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011012271046638489, -1, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:02:16.4980897Z return dequantize_per_tensor_default_1 2025-09-09T15:02:16.4981166Z 2025-09-09T15:02:16.4981440Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:16.4981802Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:02:16.4982030Z [0., 0., 0.], 2025-09-09T15:02:16.4982249Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T15:02:16.4982514Z model pt2e: GraphModule( 2025-09-09T15:02:16.4982746Z (conv): Module() 2025-09-09T15:02:16.4982940Z (bn): Module() 2025-09-09T15:02:16.4983236Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:16.4984337Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:16.4985583Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:02:16.4986084Z ) 2025-09-09T15:02:16.4986358Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:41.3837812Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:02:41.3839504Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T15:02:41.3840149Z ) 2025-09-09T15:02:41.3840487Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:41.3841880Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:41.3843502Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3977917432785034, max_val=1.4123148918151855) 2025-09-09T15:02:41.3844156Z ) 2025-09-09T15:02:41.3844358Z ) 2025-09-09T15:02:41.3844475Z 2025-09-09T15:02:41.3844480Z 2025-09-09T15:02:41.3844485Z 2025-09-09T15:02:41.3844597Z def forward(self, x): 2025-09-09T15:02:41.3844938Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:02:41.3845362Z conv_weight = self.conv.weight 2025-09-09T15:02:41.3845694Z conv_bias = self.conv.bias 2025-09-09T15:02:41.3846001Z bn_weight = self.bn.weight 2025-09-09T15:02:41.3846301Z bn_bias = self.bn.bias 2025-09-09T15:02:41.3846617Z bn_running_mean = self.bn.running_mean 2025-09-09T15:02:41.3846980Z bn_running_var = self.bn.running_var 2025-09-09T15:02:41.3847381Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:02:41.3847916Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:02:41.3849029Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:02:41.3849675Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:02:41.3850140Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:02:41.3850788Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:02:41.3851349Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:02:41.3851959Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:02:41.3852653Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:02:41.3853415Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:02:41.3854630Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:02:41.3855749Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:02:41.3856395Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:02:41.3857107Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:02:41.3857785Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:02:41.3858866Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:02:41.3860035Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:02:41.3860773Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:02:41.3861250Z 2025-09-09T15:02:41.3861591Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:41.3862031Z model fx: GraphModule( 2025-09-09T15:02:41.3862420Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:41.3864043Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:41.3865662Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:02:41.3866356Z ) 2025-09-09T15:02:41.3866563Z (conv): ConvBn1d( 2025-09-09T15:02:41.3866834Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:02:41.3867323Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:02:41.3867889Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:41.3869262Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:02:41.3870905Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T15:02:41.3871555Z ) 2025-09-09T15:02:41.3871757Z ) 2025-09-09T15:02:41.3872089Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:41.3873455Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0110], device='cuda:0'), zero_point=tensor([-1], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:41.3875239Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3977917432785034, max_val=1.4123148918151855) 2025-09-09T15:02:41.3875921Z ) 2025-09-09T15:02:41.3876142Z ) 2025-09-09T15:02:41.3876340Z 2025-09-09T15:02:41.3876353Z 2025-09-09T15:02:41.3876359Z 2025-09-09T15:02:41.3876455Z def forward(self, x): 2025-09-09T15:02:41.3878322Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:02:41.3878867Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:02:41.3879407Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:02:41.3879825Z return activation_post_process_1 2025-09-09T15:02:41.3880086Z 2025-09-09T15:02:41.3880362Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:41.3880734Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:02:41.3880960Z [0., 0., 0.], 2025-09-09T15:02:41.3881217Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T15:02:41.3881546Z converted model pt2e: GraphModule( 2025-09-09T15:02:41.3881809Z (conv): Module() 2025-09-09T15:02:41.3882014Z (bn): Module() 2025-09-09T15:02:41.3882201Z ) 2025-09-09T15:02:41.3882296Z 2025-09-09T15:02:41.3882300Z 2025-09-09T15:02:41.3882304Z 2025-09-09T15:02:41.3882393Z def forward(self, x): 2025-09-09T15:02:41.3882673Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:02:41.3883022Z conv_bias = self.conv.bias 2025-09-09T15:02:41.3883320Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:02:41.3884018Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:02:41.3885233Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:02:41.3886273Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:02:41.3886759Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:02:41.3887537Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002204145072028041, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:02:41.3888781Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:02:41.3889963Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011020027101039886, -1, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:02:41.3891251Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011020027101039886, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:02:41.3892261Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:02:41.3892664Z 2025-09-09T15:02:41.3892945Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:41.3893313Z onverted model fx: GraphModule( 2025-09-09T15:02:41.3893689Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:02:41.3894060Z ) 2025-09-09T15:02:41.3894157Z 2025-09-09T15:02:41.3894161Z 2025-09-09T15:02:41.3894165Z 2025-09-09T15:02:41.3894249Z def forward(self, x): 2025-09-09T15:02:41.3894857Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:02:41.3896064Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:02:41.3897205Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:02:41.3898047Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011020027101039886, -1, -128, 127, torch.int8); conv = None 2025-09-09T15:02:41.3899368Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011020027101039886, -1, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:02:41.3900241Z return dequantize_per_tensor_default_1 2025-09-09T15:02:41.3900512Z 2025-09-09T15:02:41.3900783Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:41.3901145Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:02:41.3901371Z [0., 0., 0.], 2025-09-09T15:02:41.3901594Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T15:02:41.3902067Z PASSED 2025-09-09T15:02:41.3902691Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion_literal_args model pt2e: GraphModule( 2025-09-09T15:02:41.3903324Z (conv): Module() 2025-09-09T15:02:41.3903529Z (bn): Module() 2025-09-09T15:02:57.6850311Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:57.6851551Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:57.6852932Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T15:02:57.6853564Z ) 2025-09-09T15:02:57.6853893Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:57.6855113Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:02:57.6856735Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T15:02:57.6857539Z ) 2025-09-09T15:02:57.6857876Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:57.6859030Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-46], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:57.6860391Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9784814715385437, max_val=2.047511577606201) 2025-09-09T15:02:57.6861027Z ) 2025-09-09T15:02:57.6861229Z ) 2025-09-09T15:02:57.6861345Z 2025-09-09T15:02:57.6861351Z 2025-09-09T15:02:57.6861370Z 2025-09-09T15:02:57.6861471Z def forward(self, x): 2025-09-09T15:02:57.6861806Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:02:57.6862219Z conv_weight = self.conv.weight 2025-09-09T15:02:57.6862548Z conv_bias = self.conv.bias 2025-09-09T15:02:57.6862856Z bn_weight = self.bn.weight 2025-09-09T15:02:57.6863166Z bn_bias = self.bn.bias 2025-09-09T15:02:57.6863470Z bn_running_mean = self.bn.running_mean 2025-09-09T15:02:57.6864084Z bn_running_var = self.bn.running_var 2025-09-09T15:02:57.6864489Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:02:57.6865038Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:02:57.6865761Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:02:57.6866406Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:02:57.6866886Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:02:57.6867811Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:02:57.6868352Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:02:57.6868962Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:02:57.6869847Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:02:57.6870596Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:02:57.6871812Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2], [4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:02:57.6872912Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:02:57.6873565Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:02:57.6874280Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:02:57.6874957Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:02:57.6876127Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:02:57.6877297Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:02:57.6878040Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:02:57.6878527Z 2025-09-09T15:02:57.6878858Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:57.6879329Z model fx: GraphModule( 2025-09-09T15:02:57.6879716Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:57.6880871Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:57.6882246Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T15:02:57.6882884Z ) 2025-09-09T15:02:57.6883095Z (conv): ConvBn1d( 2025-09-09T15:02:57.6883396Z 3, 3, kernel_size=(3,), stride=(2,), padding=(4,) 2025-09-09T15:02:57.6883918Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:02:57.6884480Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:57.6885684Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:02:57.6887326Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T15:02:57.6888150Z ) 2025-09-09T15:02:57.6888351Z ) 2025-09-09T15:02:57.6888680Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:02:57.6889837Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-46], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:02:57.6891201Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9784814715385437, max_val=2.047511577606201) 2025-09-09T15:02:57.6891839Z ) 2025-09-09T15:02:57.6892042Z ) 2025-09-09T15:02:57.6892158Z 2025-09-09T15:02:57.6892169Z 2025-09-09T15:02:57.6892174Z 2025-09-09T15:02:57.6892276Z def forward(self, x): 2025-09-09T15:02:57.6892693Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:02:57.6893453Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:02:57.6894130Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:02:57.6894645Z return activation_post_process_1 2025-09-09T15:02:57.6894963Z 2025-09-09T15:02:57.6895373Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:57.6895857Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:02:57.6896203Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:02:57.6896547Z [0., 0., 0., 0., 0., 0.]]], grad_fn=) 2025-09-09T15:02:57.6896929Z converted model pt2e: GraphModule( 2025-09-09T15:02:57.6897248Z (conv): Module() 2025-09-09T15:02:57.6897494Z (bn): Module() 2025-09-09T15:02:57.6897720Z ) 2025-09-09T15:02:57.6897835Z 2025-09-09T15:02:57.6897840Z 2025-09-09T15:02:57.6897845Z 2025-09-09T15:02:57.6897952Z def forward(self, x): 2025-09-09T15:02:57.6898290Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:02:57.6898694Z conv_bias = self.conv.bias 2025-09-09T15:02:57.6899053Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:02:57.6899935Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T15:02:57.6901482Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:02:57.6902767Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:02:57.6903342Z _scale_0 = self._scale_0 2025-09-09T15:02:57.6903645Z _zero_point_0 = self._zero_point_0 2025-09-09T15:02:57.6904013Z quantize_per_channel = self._frozen_param0 2025-09-09T15:02:57.6905170Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:02:57.6906822Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias, [2], [4]); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:02:57.6908026Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011866639368236065, -46, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:02:57.6909314Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011866639368236065, -46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:02:57.6910314Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:02:57.6910735Z 2025-09-09T15:02:57.6911009Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:02:57.6911384Z onverted model fx: GraphModule( 2025-09-09T15:02:57.6911791Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(2,), padding=(4,)) 2025-09-09T15:02:57.6912203Z ) 2025-09-09T15:02:57.6912301Z 2025-09-09T15:02:57.6912309Z 2025-09-09T15:02:57.6912313Z 2025-09-09T15:02:57.6912405Z def forward(self, x): 2025-09-09T15:02:57.6913018Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T15:02:57.6914252Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:02:57.6915255Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:02:57.6916269Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011866639368236065, -46, -128, 127, torch.int8); conv = None 2025-09-09T15:02:57.6917621Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011866639368236065, -46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:03:16.9148252Z return dequantize_per_tensor_default_1 2025-09-09T15:03:16.9148682Z 2025-09-09T15:03:16.9149020Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:16.9149482Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:03:16.9149808Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:03:16.9150104Z [0., 0., 0., 0., 0., 0.]]]) 2025-09-09T15:03:16.9150420Z model pt2e: GraphModule( 2025-09-09T15:03:16.9150692Z (conv): Module() 2025-09-09T15:03:16.9150939Z (bn): Module() 2025-09-09T15:03:16.9151293Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:16.9152504Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:16.9153917Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T15:03:16.9154549Z ) 2025-09-09T15:03:16.9154884Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:16.9156169Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:03:16.9157575Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.31192728877067566, max_val=0.23078496754169464) 2025-09-09T15:03:16.9158247Z ) 2025-09-09T15:03:16.9158607Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:16.9159773Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-44], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:16.9161144Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9980934262275696, max_val=2.047511577606201) 2025-09-09T15:03:16.9161784Z ) 2025-09-09T15:03:16.9161989Z ) 2025-09-09T15:03:16.9162115Z 2025-09-09T15:03:16.9162120Z 2025-09-09T15:03:16.9162124Z 2025-09-09T15:03:16.9162226Z def forward(self, x): 2025-09-09T15:03:16.9162565Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:03:16.9162968Z conv_weight = self.conv.weight 2025-09-09T15:03:16.9163297Z conv_bias = self.conv.bias 2025-09-09T15:03:16.9163598Z bn_weight = self.bn.weight 2025-09-09T15:03:16.9164120Z bn_bias = self.bn.bias 2025-09-09T15:03:16.9164432Z bn_running_mean = self.bn.running_mean 2025-09-09T15:03:16.9164796Z bn_running_var = self.bn.running_var 2025-09-09T15:03:16.9165196Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:03:16.9165731Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:03:16.9166463Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:03:16.9167103Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:03:16.9167583Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:03:16.9168072Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:03:16.9168608Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:03:16.9169218Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:03:16.9169900Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:03:16.9171065Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:03:16.9172442Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2], [4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:03:16.9173559Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:03:16.9174212Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:03:16.9174910Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:03:16.9175595Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:03:16.9176665Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:03:16.9177838Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:03:16.9178579Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:03:16.9179092Z 2025-09-09T15:03:16.9179435Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:16.9179868Z model fx: GraphModule( 2025-09-09T15:03:16.9180374Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:16.9181889Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([57], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:16.9183370Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.611703872680664, max_val=0.6104744076728821) 2025-09-09T15:03:16.9192911Z ) 2025-09-09T15:03:16.9193177Z (conv): ConvBn1d( 2025-09-09T15:03:16.9193487Z 3, 3, kernel_size=(3,), stride=(2,), padding=(4,) 2025-09-09T15:03:16.9194017Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:03:16.9194592Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:16.9195730Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:03:16.9197292Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.31192728877067566, max_val=0.23078496754169464) 2025-09-09T15:03:16.9197959Z ) 2025-09-09T15:03:16.9198190Z ) 2025-09-09T15:03:16.9198560Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:16.9199720Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0119]), zero_point=tensor([-44], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:16.9201101Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9980934262275696, max_val=2.047511577606201) 2025-09-09T15:03:16.9201742Z ) 2025-09-09T15:03:16.9201948Z ) 2025-09-09T15:03:16.9202084Z 2025-09-09T15:03:16.9202103Z 2025-09-09T15:03:16.9202109Z 2025-09-09T15:03:16.9202226Z def forward(self, x): 2025-09-09T15:03:16.9202690Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:03:16.9203402Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:03:16.9204027Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:03:16.9204455Z return activation_post_process_1 2025-09-09T15:03:16.9204728Z 2025-09-09T15:03:16.9205011Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:16.9205539Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:03:16.9205814Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:03:16.9206105Z [0., 0., 0., 0., 0., 0.]]], grad_fn=) 2025-09-09T15:03:16.9206421Z converted model pt2e: GraphModule( 2025-09-09T15:03:16.9206689Z (conv): Module() 2025-09-09T15:03:16.9206978Z (bn): Module() 2025-09-09T15:03:16.9207180Z ) 2025-09-09T15:03:16.9207282Z 2025-09-09T15:03:16.9207286Z 2025-09-09T15:03:16.9207290Z 2025-09-09T15:03:16.9207377Z def forward(self, x): 2025-09-09T15:03:16.9207669Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:03:16.9208006Z conv_bias = self.conv.bias 2025-09-09T15:03:16.9208303Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:03:16.9209014Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T15:03:16.9210245Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:03:16.9211295Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:03:16.9211795Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:03:16.9212588Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0024561204481869936, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:03:16.9213853Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias, [2], [4]); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:03:16.9215047Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.011943548917770386, -44, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:03:16.9216347Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011943548917770386, -44, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:03:16.9217359Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:03:16.9217772Z 2025-09-09T15:03:16.9218059Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:16.9218437Z onverted model fx: GraphModule( 2025-09-09T15:03:16.9218901Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(2,), padding=(4,)) 2025-09-09T15:03:16.9219321Z ) 2025-09-09T15:03:16.9219423Z 2025-09-09T15:03:16.9219427Z 2025-09-09T15:03:16.9219431Z 2025-09-09T15:03:16.9219518Z def forward(self, x): 2025-09-09T15:03:16.9220139Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008714424446225166, 57, -128, 127, torch.int8); x = None 2025-09-09T15:03:16.9221368Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008714424446225166, 57, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:03:16.9222381Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:03:16.9223241Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.011943548917770386, -44, -128, 127, torch.int8); conv = None 2025-09-09T15:03:43.4328582Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.011943548917770386, -44, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:03:43.4330224Z return dequantize_per_tensor_default_1 2025-09-09T15:03:43.4330757Z 2025-09-09T15:03:43.4331148Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:43.4334468Z diff: tensor([[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:03:43.4334767Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:03:43.4335145Z [0., 0., 0., 0., 0., 0.]]]) 2025-09-09T15:03:43.4335734Z PASSED 2025-09-09T15:03:43.4336586Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T15:03:43.4337237Z (conv): Module() 2025-09-09T15:03:43.4337440Z (bn): Module() 2025-09-09T15:03:43.4337739Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:43.4338674Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:43.4339888Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:03:43.4340585Z ) 2025-09-09T15:03:43.4340895Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:43.4342257Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0020, 0.0022]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:03:43.4343554Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618]), max_val=tensor([0.1970, 0.2308, 0.2775])) 2025-09-09T15:03:43.4344194Z ) 2025-09-09T15:03:43.4344467Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:43.4345393Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:43.4346464Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T15:03:43.4346975Z ) 2025-09-09T15:03:43.4347148Z ) 2025-09-09T15:03:43.4347243Z 2025-09-09T15:03:43.4347247Z 2025-09-09T15:03:43.4347251Z 2025-09-09T15:03:43.4347336Z def forward(self, x): 2025-09-09T15:03:43.4347628Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:03:43.4347963Z conv_weight = self.conv.weight 2025-09-09T15:03:43.4348240Z bn_weight = self.bn.weight 2025-09-09T15:03:43.4348487Z bn_bias = self.bn.bias 2025-09-09T15:03:43.4348744Z bn_running_mean = self.bn.running_mean 2025-09-09T15:03:43.4349040Z bn_running_var = self.bn.running_var 2025-09-09T15:03:43.4349376Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:03:43.4349814Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:03:43.4350397Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:03:43.4350933Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:03:43.4351321Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:03:43.4351740Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:03:43.4352180Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:03:43.4352683Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:03:43.4353246Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:03:43.4354075Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:03:43.4354893Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:03:43.4355419Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:03:43.4356569Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:03:43.4357594Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:03:43.4358192Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:03:43.4358587Z 2025-09-09T15:03:43.4358867Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:43.4359239Z model fx: GraphModule( 2025-09-09T15:03:43.4359570Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:43.4360511Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:43.4361628Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:03:43.4362142Z ) 2025-09-09T15:03:43.4362329Z (conv): ConvBn1d( 2025-09-09T15:03:43.4362564Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:03:43.4363001Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:03:43.4363469Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:43.4364742Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0020, 0.0022]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:03:43.4366037Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618]), max_val=tensor([0.1970, 0.2308, 0.2775])) 2025-09-09T15:03:43.4366726Z ) 2025-09-09T15:03:43.4366906Z ) 2025-09-09T15:03:43.4367183Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:03:43.4368116Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:03:43.4369211Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T15:03:43.4369714Z ) 2025-09-09T15:03:43.4369885Z ) 2025-09-09T15:03:43.4369978Z 2025-09-09T15:03:43.4370002Z 2025-09-09T15:03:43.4370006Z 2025-09-09T15:03:43.4370097Z def forward(self, x): 2025-09-09T15:03:43.4370443Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:03:43.4370978Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:03:43.4371521Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:03:43.4371960Z return activation_post_process_1 2025-09-09T15:03:43.4372223Z 2025-09-09T15:03:43.4372501Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:43.4372869Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:03:43.4373101Z [0., 0., 0.], 2025-09-09T15:03:43.4373317Z [0., 0., 0.]], 2025-09-09T15:03:43.4373451Z 2025-09-09T15:03:43.4373527Z [[0., 0., 0.], 2025-09-09T15:03:43.4373733Z [0., 0., 0.], 2025-09-09T15:03:43.4373932Z [0., 0., 0.]], 2025-09-09T15:03:43.4374075Z 2025-09-09T15:03:43.4374147Z [[0., 0., 0.], 2025-09-09T15:03:43.4374354Z [0., 0., 0.], 2025-09-09T15:03:43.4374581Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:03:43.4374889Z converted model pt2e: GraphModule( 2025-09-09T15:03:43.4375149Z (conv): Module() 2025-09-09T15:03:43.4375355Z (bn): Module() 2025-09-09T15:03:43.4375543Z ) 2025-09-09T15:03:43.4375644Z 2025-09-09T15:03:43.4375809Z 2025-09-09T15:03:43.4375813Z 2025-09-09T15:03:43.4375898Z def forward(self, x): 2025-09-09T15:03:43.4376177Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:03:43.4376556Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:03:43.4377382Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:03:43.4378616Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:03:43.4379658Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:03:43.4380122Z _scale_0 = self._scale_0 2025-09-09T15:03:43.4380386Z _zero_point_0 = self._zero_point_0 2025-09-09T15:03:43.4380695Z quantize_per_channel = self._frozen_param0 2025-09-09T15:03:43.4381581Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:03:43.4382467Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:03:43.4383312Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T15:03:43.4384564Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.014605328440666199, 8, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:03:43.4385855Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:03:43.4386860Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:03:43.4387275Z 2025-09-09T15:03:43.4387557Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:03:43.4387930Z onverted model fx: GraphModule( 2025-09-09T15:03:43.4388308Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:03:43.4388673Z ) 2025-09-09T15:03:43.4388767Z 2025-09-09T15:03:43.4388778Z 2025-09-09T15:03:43.4388782Z 2025-09-09T15:03:43.4388869Z def forward(self, x): 2025-09-09T15:03:43.4389480Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:03:43.4390710Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:04:02.7016933Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:04:02.7018203Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.014605328440666199, 8, -128, 127, torch.int8); conv = None 2025-09-09T15:04:02.7019517Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:04:02.7020382Z return dequantize_per_tensor_default_1 2025-09-09T15:04:02.7020663Z 2025-09-09T15:04:02.7020947Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:02.7021321Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:04:02.7021566Z [0., 0., 0.], 2025-09-09T15:04:02.7021771Z [0., 0., 0.]], 2025-09-09T15:04:02.7021904Z 2025-09-09T15:04:02.7021986Z [[0., 0., 0.], 2025-09-09T15:04:02.7022186Z [0., 0., 0.], 2025-09-09T15:04:02.7022784Z [0., 0., 0.]], 2025-09-09T15:04:02.7022928Z 2025-09-09T15:04:02.7023001Z [[0., 0., 0.], 2025-09-09T15:04:02.7023203Z [0., 0., 0.], 2025-09-09T15:04:02.7023406Z [0., 0., 0.]]]) 2025-09-09T15:04:02.7023631Z model pt2e: GraphModule( 2025-09-09T15:04:02.7023860Z (conv): Module() 2025-09-09T15:04:02.7024262Z (bn): Module() 2025-09-09T15:04:02.7024575Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:02.7025498Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:02.7026610Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:04:02.7027126Z ) 2025-09-09T15:04:02.7027402Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:02.7028345Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:04:02.7029449Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T15:04:02.7029954Z ) 2025-09-09T15:04:02.7030231Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:02.7031145Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:02.7032217Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T15:04:02.7032716Z ) 2025-09-09T15:04:02.7032891Z ) 2025-09-09T15:04:02.7032991Z 2025-09-09T15:04:02.7032995Z 2025-09-09T15:04:02.7032999Z 2025-09-09T15:04:02.7033085Z def forward(self, x): 2025-09-09T15:04:02.7033364Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:04:02.7033702Z conv_weight = self.conv.weight 2025-09-09T15:04:02.7033967Z bn_weight = self.bn.weight 2025-09-09T15:04:02.7034225Z bn_bias = self.bn.bias 2025-09-09T15:04:02.7034474Z bn_running_mean = self.bn.running_mean 2025-09-09T15:04:02.7034770Z bn_running_var = self.bn.running_var 2025-09-09T15:04:02.7035097Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:04:02.7035531Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:04:02.7036114Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:04:02.7036727Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:04:02.7037119Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:04:02.7037526Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:04:02.7037962Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:04:02.7038477Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:04:02.7039066Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:04:02.7039893Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:04:02.7040699Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:04:02.7041224Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:04:02.7042100Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:04:02.7043099Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:04:02.7043690Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:04:02.7044070Z 2025-09-09T15:04:02.7044421Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:02.7044778Z model fx: GraphModule( 2025-09-09T15:04:02.7045106Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:02.7046037Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:02.7047120Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:04:02.7047628Z ) 2025-09-09T15:04:02.7047805Z (conv): ConvBn1d( 2025-09-09T15:04:02.7048045Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:04:02.7048463Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:04:02.7048971Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:02.7049882Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:04:02.7050977Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T15:04:02.7051493Z ) 2025-09-09T15:04:02.7051656Z ) 2025-09-09T15:04:02.7051929Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:02.7052856Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0146]), zero_point=tensor([8], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:02.7053933Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9912875890731812, max_val=1.733071208000183) 2025-09-09T15:04:02.7054435Z ) 2025-09-09T15:04:02.7054599Z ) 2025-09-09T15:04:02.7054704Z 2025-09-09T15:04:02.7054709Z 2025-09-09T15:04:02.7054713Z 2025-09-09T15:04:02.7054796Z def forward(self, x): 2025-09-09T15:04:02.7055145Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:04:02.7055668Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:04:02.7056208Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:04:02.7056624Z return activation_post_process_1 2025-09-09T15:04:02.7056882Z 2025-09-09T15:04:02.7057151Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:02.7057517Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:04:02.7057740Z [0., 0., 0.], 2025-09-09T15:04:02.7057973Z [0., 0., 0.]], 2025-09-09T15:04:02.7058130Z 2025-09-09T15:04:02.7058210Z [[0., 0., 0.], 2025-09-09T15:04:02.7058405Z [0., 0., 0.], 2025-09-09T15:04:02.7058609Z [0., 0., 0.]], 2025-09-09T15:04:02.7058742Z 2025-09-09T15:04:02.7058815Z [[0., 0., 0.], 2025-09-09T15:04:02.7059014Z [0., 0., 0.], 2025-09-09T15:04:02.7059238Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:04:02.7059542Z converted model pt2e: GraphModule( 2025-09-09T15:04:02.7059795Z (conv): Module() 2025-09-09T15:04:02.7059996Z (bn): Module() 2025-09-09T15:04:02.7060189Z ) 2025-09-09T15:04:02.7060282Z 2025-09-09T15:04:02.7060286Z 2025-09-09T15:04:02.7060289Z 2025-09-09T15:04:02.7060373Z def forward(self, x): 2025-09-09T15:04:02.7060649Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:04:02.7061014Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:04:02.7061802Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:04:02.7063097Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:04:02.7064565Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:04:02.7065052Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:04:02.7065825Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002204145072028041, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:04:02.7066611Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:04:02.7067434Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T15:04:02.7068670Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.014605328440666199, 8, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:04:02.7069940Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:04:02.7070931Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:04:02.7071334Z 2025-09-09T15:04:02.7071613Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:02.7071980Z onverted model fx: GraphModule( 2025-09-09T15:04:02.7072351Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:04:02.7072719Z ) 2025-09-09T15:04:02.7072828Z 2025-09-09T15:04:02.7072832Z 2025-09-09T15:04:02.7072836Z 2025-09-09T15:04:02.7072920Z def forward(self, x): 2025-09-09T15:04:02.7073528Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:04:02.7074742Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:04:05.6648727Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:04:05.6649758Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.014605328440666199, 8, -128, 127, torch.int8); conv = None 2025-09-09T15:04:05.6651257Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.014605328440666199, 8, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:04:05.6652405Z return dequantize_per_tensor_default_1 2025-09-09T15:04:05.6652782Z 2025-09-09T15:04:05.6653151Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:05.6653616Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:04:05.6653851Z [0., 0., 0.], 2025-09-09T15:04:05.6654061Z [0., 0., 0.]], 2025-09-09T15:04:05.6654196Z 2025-09-09T15:04:05.6654273Z [[0., 0., 0.], 2025-09-09T15:04:05.6654482Z [0., 0., 0.], 2025-09-09T15:04:05.6654680Z [0., 0., 0.]], 2025-09-09T15:04:05.6654820Z 2025-09-09T15:04:05.6654892Z [[0., 0., 0.], 2025-09-09T15:04:05.6655092Z [0., 0., 0.], 2025-09-09T15:04:05.6655290Z [0., 0., 0.]]]) 2025-09-09T15:04:05.6655519Z model pt2e: GraphModule( 2025-09-09T15:04:05.6655743Z (conv1): Module() 2025-09-09T15:04:05.6655944Z (bn1): Module() 2025-09-09T15:04:05.6656503Z (conv2): Module() 2025-09-09T15:04:05.6656705Z (bn2): Module() 2025-09-09T15:04:05.6657004Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6658089Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:05.6659203Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:04:05.6659707Z ) 2025-09-09T15:04:05.6659987Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6661011Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0023, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:04:05.6662296Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2878, -0.2584, -0.3162]), max_val=tensor([0.2745, 0.3315, 0.3105])) 2025-09-09T15:04:05.6662942Z ) 2025-09-09T15:04:05.6663213Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6664387Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0023]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:04:05.6665654Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3233, -0.3086, -0.2979]), max_val=tensor([0.3026, 0.1712, 0.2405])) 2025-09-09T15:04:05.6666292Z ) 2025-09-09T15:04:05.6666569Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6667491Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:05.6668584Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.805302619934082, max_val=2.310788631439209) 2025-09-09T15:04:05.6669080Z ) 2025-09-09T15:04:05.6669360Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6670282Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:05.6671366Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.1982380151748657, max_val=1.4133442640304565) 2025-09-09T15:04:05.6671876Z ) 2025-09-09T15:04:05.6672051Z ) 2025-09-09T15:04:05.6672159Z 2025-09-09T15:04:05.6672164Z 2025-09-09T15:04:05.6672168Z 2025-09-09T15:04:05.6672251Z def forward(self, x): 2025-09-09T15:04:05.6672541Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:04:05.6672877Z conv1_weight = self.conv1.weight 2025-09-09T15:04:05.6673167Z bn1_weight = self.bn1.weight 2025-09-09T15:04:05.6673429Z bn1_bias = self.bn1.bias 2025-09-09T15:04:05.6673688Z conv2_weight = self.conv2.weight 2025-09-09T15:04:05.6673962Z conv2_bias = self.conv2.bias 2025-09-09T15:04:05.6674226Z bn2_weight = self.bn2.weight 2025-09-09T15:04:05.6674474Z bn2_bias = self.bn2.bias 2025-09-09T15:04:05.6674744Z bn1_running_mean = self.bn1.running_mean 2025-09-09T15:04:05.6675056Z bn1_running_var = self.bn1.running_var 2025-09-09T15:04:05.6675394Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T15:04:05.6675744Z bn2_running_mean = self.bn2.running_mean 2025-09-09T15:04:05.6676041Z bn2_running_var = self.bn2.running_var 2025-09-09T15:04:05.6676468Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T15:04:05.6677037Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:04:05.6677628Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T15:04:05.6678293Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T15:04:05.6678925Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T15:04:05.6679329Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:04:05.6679736Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T15:04:05.6680176Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:04:05.6680673Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T15:04:05.6681239Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T15:04:05.6681853Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:04:05.6682401Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T15:04:05.6682806Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T15:04:05.6683235Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T15:04:05.6683701Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1]) 2025-09-09T15:04:05.6684220Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T15:04:05.6684809Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T15:04:05.6685651Z conv1d_3 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:04:05.6686467Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1]); div_2 = None 2025-09-09T15:04:05.6687004Z div_3 = torch.ops.aten.div.Tensor(conv1d_3, reshape_4); conv1d_3 = reshape_4 = None 2025-09-09T15:04:05.6687917Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T15:04:05.6688871Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T15:04:05.6689820Z conv1d_2 = torch.ops.aten.conv1d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T15:04:05.6690685Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:04:05.6691216Z div_1 = torch.ops.aten.div.Tensor(conv1d_2, reshape_1); conv1d_2 = reshape_1 = None 2025-09-09T15:04:05.6691788Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1]); conv2_bias = None 2025-09-09T15:04:05.6692337Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:04:05.6693227Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T15:04:05.6694172Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T15:04:05.6694762Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T15:04:05.6695144Z 2025-09-09T15:04:05.6695421Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:05.6695786Z model fx: GraphModule( 2025-09-09T15:04:05.6696105Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6697040Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:05.6698223Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:04:05.6706723Z ) 2025-09-09T15:04:05.6706942Z (conv1): ConvBn1d( 2025-09-09T15:04:05.6707325Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:04:05.6707761Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:04:05.6708242Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6709207Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0023]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:04:05.6710573Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3233, -0.3086, -0.2979]), max_val=tensor([0.3026, 0.1712, 0.2405])) 2025-09-09T15:04:05.6711236Z ) 2025-09-09T15:04:05.6711409Z ) 2025-09-09T15:04:05.6711697Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:05.6712652Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:05.6713749Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.805302619934082, max_val=2.310788631439209) 2025-09-09T15:04:05.6714259Z ) 2025-09-09T15:04:05.6714439Z (conv2): ConvBn1d( 2025-09-09T15:04:25.1248736Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:04:25.1249356Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:04:25.1249953Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:25.1251200Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0023, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:04:25.1252929Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2878, -0.2584, -0.3162]), max_val=tensor([0.2745, 0.3315, 0.3105])) 2025-09-09T15:04:25.1253751Z ) 2025-09-09T15:04:25.1253964Z ) 2025-09-09T15:04:25.1254296Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:25.1255486Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:25.1256889Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.1982380151748657, max_val=1.4133442640304565) 2025-09-09T15:04:25.1257543Z ) 2025-09-09T15:04:25.1257755Z ) 2025-09-09T15:04:25.1257875Z 2025-09-09T15:04:25.1257880Z 2025-09-09T15:04:25.1257885Z 2025-09-09T15:04:25.1257989Z def forward(self, x): 2025-09-09T15:04:25.1258419Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:04:25.1259086Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:04:25.1259782Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T15:04:25.1260473Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:04:25.1261150Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T15:04:25.1261682Z return activation_post_process_2 2025-09-09T15:04:25.1261998Z 2025-09-09T15:04:25.1262337Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:25.1262818Z diff: tensor([[[0.], 2025-09-09T15:04:25.1263387Z [0.], 2025-09-09T15:04:25.1263610Z [0.]], 2025-09-09T15:04:25.1264055Z 2025-09-09T15:04:25.1264154Z [[0.], 2025-09-09T15:04:25.1264383Z [0.], 2025-09-09T15:04:25.1264611Z [0.]], 2025-09-09T15:04:25.1264749Z 2025-09-09T15:04:25.1264846Z [[0.], 2025-09-09T15:04:25.1265064Z [0.], 2025-09-09T15:04:25.1265494Z [0.]]], grad_fn=) 2025-09-09T15:04:25.1265849Z converted model pt2e: GraphModule( 2025-09-09T15:04:25.1266175Z (conv1): Module() 2025-09-09T15:04:25.1266419Z (bn1): Module() 2025-09-09T15:04:25.1266665Z (conv2): Module() 2025-09-09T15:04:25.1266907Z (bn2): Module() 2025-09-09T15:04:25.1267145Z ) 2025-09-09T15:04:25.1267262Z 2025-09-09T15:04:25.1267267Z 2025-09-09T15:04:25.1267272Z 2025-09-09T15:04:25.1267380Z def forward(self, x): 2025-09-09T15:04:25.1267713Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:04:25.1268125Z conv2_bias = self.conv2.bias 2025-09-09T15:04:25.1268535Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T15:04:25.1269014Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T15:04:25.1269902Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:04:25.1271459Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:04:25.1272840Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T15:04:25.1273666Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T15:04:25.1274258Z _scale_0 = self._scale_0 2025-09-09T15:04:25.1274565Z _zero_point_0 = self._zero_point_0 2025-09-09T15:04:25.1274905Z _scale_1 = self._scale_1 2025-09-09T15:04:25.1275224Z _zero_point_1 = self._zero_point_1 2025-09-09T15:04:25.1275591Z quantize_per_channel_1 = self._frozen_param0 2025-09-09T15:04:25.1276823Z dequantize_per_channel_1 = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_1, _scale_1, _zero_point_1, 0, -127, 127, torch.int8); quantize_per_channel_1 = _scale_1 = _zero_point_1 = None 2025-09-09T15:04:25.1277962Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T15:04:25.1279051Z conv1d_5 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_channel_1 = conv1_weight_bias = None 2025-09-09T15:04:25.1280698Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_5, 0.016141533851623535, -16, -128, 127, torch.int8); conv1d_5 = None 2025-09-09T15:04:25.1282356Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016141533851623535, -16, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:04:25.1283510Z quantize_per_channel = self._frozen_param1 2025-09-09T15:04:25.1284670Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:04:25.1286021Z conv1d_4 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_1, dequantize_per_channel, conv2_bias); dequantize_per_tensor_default_1 = dequantize_per_channel = conv2_bias = None 2025-09-09T15:04:25.1287228Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_4, 0.010241499170660973, -11, -128, 127, torch.int8); conv1d_4 = None 2025-09-09T15:04:25.1288509Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010241499170660973, -11, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:04:25.1289657Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:04:25.1290072Z 2025-09-09T15:04:25.1290353Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:25.1290813Z onverted model fx: GraphModule( 2025-09-09T15:04:25.1291186Z (conv1): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:04:25.1291690Z (conv2): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:04:25.1292062Z ) 2025-09-09T15:04:25.1292161Z 2025-09-09T15:04:25.1292165Z 2025-09-09T15:04:25.1292169Z 2025-09-09T15:04:25.1292255Z def forward(self, x): 2025-09-09T15:04:25.1292919Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:04:25.1294142Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:04:25.1295158Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:04:25.1296027Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.016141533851623535, -16, -128, 127, torch.int8); conv1 = None 2025-09-09T15:04:25.1297290Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016141533851623535, -16, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:04:25.1298320Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:04:25.1299188Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.010241499170660973, -11, -128, 127, torch.int8); conv2 = None 2025-09-09T15:04:25.1300452Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010241499170660973, -11, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:04:25.1301327Z return dequantize_per_tensor_default_2 2025-09-09T15:04:25.1301591Z 2025-09-09T15:04:25.1301870Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:25.1302234Z diff: tensor([[[0.], 2025-09-09T15:04:25.1302442Z [0.], 2025-09-09T15:04:25.1302635Z [0.]], 2025-09-09T15:04:25.1302751Z 2025-09-09T15:04:25.1302825Z [[0.], 2025-09-09T15:04:25.1303013Z [0.], 2025-09-09T15:04:25.1303192Z [0.]], 2025-09-09T15:04:25.1303314Z 2025-09-09T15:04:25.1303386Z [[0.], 2025-09-09T15:04:25.1303566Z [0.], 2025-09-09T15:04:25.1303752Z [0.]]]) 2025-09-09T15:04:25.1303960Z model pt2e: GraphModule( 2025-09-09T15:04:25.1304198Z (conv1): Module() 2025-09-09T15:04:25.1304398Z (bn1): Module() 2025-09-09T15:04:25.1304593Z (conv2): Module() 2025-09-09T15:04:25.1304796Z (bn2): Module() 2025-09-09T15:04:25.1305094Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:25.1306030Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:25.1307114Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:04:25.1307626Z ) 2025-09-09T15:04:25.1307907Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:25.1308838Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:04:25.1310031Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3161814510822296, max_val=0.33154603838920593) 2025-09-09T15:04:25.1310538Z ) 2025-09-09T15:04:25.1310816Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:25.1311834Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:04:25.1313169Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232710361480713, max_val=0.30256387591362) 2025-09-09T15:04:25.1313797Z ) 2025-09-09T15:04:25.1314137Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6839299Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:41.6840506Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.807204008102417, max_val=2.3096539974212646) 2025-09-09T15:04:41.6841022Z ) 2025-09-09T15:04:41.6841326Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6842252Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:41.6843395Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.2001533508300781, max_val=1.4126498699188232) 2025-09-09T15:04:41.6843904Z ) 2025-09-09T15:04:41.6844073Z ) 2025-09-09T15:04:41.6844172Z 2025-09-09T15:04:41.6844176Z 2025-09-09T15:04:41.6844180Z 2025-09-09T15:04:41.6844271Z def forward(self, x): 2025-09-09T15:04:41.6844556Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:04:41.6844895Z conv1_weight = self.conv1.weight 2025-09-09T15:04:41.6845176Z bn1_weight = self.bn1.weight 2025-09-09T15:04:41.6845438Z bn1_bias = self.bn1.bias 2025-09-09T15:04:41.6845685Z conv2_weight = self.conv2.weight 2025-09-09T15:04:41.6845965Z conv2_bias = self.conv2.bias 2025-09-09T15:04:41.6846229Z bn2_weight = self.bn2.weight 2025-09-09T15:04:41.6846476Z bn2_bias = self.bn2.bias 2025-09-09T15:04:41.6846753Z bn1_running_mean = self.bn1.running_mean 2025-09-09T15:04:41.6847055Z bn1_running_var = self.bn1.running_var 2025-09-09T15:04:41.6847393Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T15:04:41.6847736Z bn2_running_mean = self.bn2.running_mean 2025-09-09T15:04:41.6848038Z bn2_running_var = self.bn2.running_var 2025-09-09T15:04:41.6848366Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T15:04:41.6848815Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:04:41.6849416Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T15:04:41.6850077Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T15:04:41.6850621Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T15:04:41.6851012Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:04:41.6851432Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T15:04:41.6851876Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:04:41.6852384Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T15:04:41.6852954Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T15:04:41.6853562Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:04:41.6854519Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T15:04:41.6854926Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T15:04:41.6855362Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T15:04:41.6856034Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1]) 2025-09-09T15:04:41.6856566Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T15:04:41.6857161Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T15:04:41.6858003Z conv1d_3 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:04:41.6858838Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1]); div_2 = None 2025-09-09T15:04:41.6859375Z div_3 = torch.ops.aten.div.Tensor(conv1d_3, reshape_4); conv1d_3 = reshape_4 = None 2025-09-09T15:04:41.6860307Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T15:04:41.6861269Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T15:04:41.6862213Z conv1d_2 = torch.ops.aten.conv1d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T15:04:41.6863092Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:04:41.6863620Z div_1 = torch.ops.aten.div.Tensor(conv1d_2, reshape_1); conv1d_2 = reshape_1 = None 2025-09-09T15:04:41.6864544Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1]); conv2_bias = None 2025-09-09T15:04:41.6865104Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:04:41.6865984Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T15:04:41.6866934Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T15:04:41.6867530Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T15:04:41.6867911Z 2025-09-09T15:04:41.6868192Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:41.6868554Z model fx: GraphModule( 2025-09-09T15:04:41.6868875Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6869800Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0150]), zero_point=tensor([14], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:41.6870913Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1280412673950195, max_val=1.6863642930984497) 2025-09-09T15:04:41.6871415Z ) 2025-09-09T15:04:41.6871596Z (conv1): ConvBn1d( 2025-09-09T15:04:41.6871845Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:04:41.6872260Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:04:41.6872719Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6873673Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:04:41.6874764Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232710361480713, max_val=0.30256387591362) 2025-09-09T15:04:41.6875425Z ) 2025-09-09T15:04:41.6875595Z ) 2025-09-09T15:04:41.6875870Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6877036Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-16], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:41.6878127Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.807204008102417, max_val=2.3096539974212646) 2025-09-09T15:04:41.6878627Z ) 2025-09-09T15:04:41.6878808Z (conv2): ConvBn1d( 2025-09-09T15:04:41.6879034Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:04:41.6879437Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:04:41.6879896Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6880799Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:04:41.6881910Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3161814510822296, max_val=0.33154603838920593) 2025-09-09T15:04:41.6882415Z ) 2025-09-09T15:04:41.6882590Z ) 2025-09-09T15:04:41.6882879Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:04:41.6883849Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0102]), zero_point=tensor([-11], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:04:41.6884935Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.2001533508300781, max_val=1.4126498699188232) 2025-09-09T15:04:41.6885434Z ) 2025-09-09T15:04:41.6885610Z ) 2025-09-09T15:04:41.6885704Z 2025-09-09T15:04:41.6885709Z 2025-09-09T15:04:41.6885717Z 2025-09-09T15:04:41.6885803Z def forward(self, x): 2025-09-09T15:04:41.6886148Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:04:41.6886685Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:04:41.6887233Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T15:04:41.6887788Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:04:41.6888327Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T15:04:41.6888756Z return activation_post_process_2 2025-09-09T15:04:41.6889017Z 2025-09-09T15:04:41.6889286Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:04:41.6889640Z diff: tensor([[[0.], 2025-09-09T15:04:41.6889847Z [0.], 2025-09-09T15:04:41.6890035Z [0.]], 2025-09-09T15:04:41.6890151Z 2025-09-09T15:04:41.6890222Z [[0.], 2025-09-09T15:04:41.6890410Z [0.], 2025-09-09T15:04:41.6890589Z [0.]], 2025-09-09T15:04:41.6890710Z 2025-09-09T15:04:41.6890782Z [[0.], 2025-09-09T15:04:41.6890967Z [0.], 2025-09-09T15:04:41.6891177Z [0.]]], grad_fn=) 2025-09-09T15:04:41.6891472Z converted model pt2e: GraphModule( 2025-09-09T15:04:41.6891736Z (conv1): Module() 2025-09-09T15:04:41.6891940Z (bn1): Module() 2025-09-09T15:04:41.6892135Z (conv2): Module() 2025-09-09T15:04:41.6892333Z (bn2): Module() 2025-09-09T15:04:41.6892517Z ) 2025-09-09T15:04:41.6892619Z 2025-09-09T15:04:41.6892623Z 2025-09-09T15:04:41.6892627Z 2025-09-09T15:04:41.6892711Z def forward(self, x): 2025-09-09T15:04:41.6892984Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:04:41.6893319Z conv2_bias = self.conv2.bias 2025-09-09T15:04:41.6893636Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T15:04:41.6894004Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T15:04:41.6894810Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:04:41.6896107Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:05:08.3366400Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T15:05:08.3368812Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T15:05:08.3369468Z quantize_per_tensor_1 = self._frozen_param0 2025-09-09T15:05:08.3370476Z dequantize_per_tensor_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_1, 0.0025454412680119276, 0, -127, 127, torch.int8); quantize_per_tensor_1 = None 2025-09-09T15:05:08.3371503Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T15:05:08.3372551Z conv1d_5 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor_1 = conv1_weight_bias = None 2025-09-09T15:05:08.3374123Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_5, 0.016144542023539543, -16, -128, 127, torch.int8); conv1d_5 = None 2025-09-09T15:05:08.3375732Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.016144542023539543, -16, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T15:05:08.3376820Z quantize_per_tensor = self._frozen_param1 2025-09-09T15:05:08.3377792Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0026105986908078194, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:05:08.3379370Z conv1d_4 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_3, dequantize_per_tensor, conv2_bias); dequantize_per_tensor_default_3 = dequantize_per_tensor = conv2_bias = None 2025-09-09T15:05:08.3380907Z quantize_per_tensor_default_4 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_4, 0.01024628710001707, -11, -128, 127, torch.int8); conv1d_4 = None 2025-09-09T15:05:08.3382495Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_4, 0.01024628710001707, -11, -128, 127, torch.int8); quantize_per_tensor_default_4 = None 2025-09-09T15:05:08.3383731Z return pytree.tree_unflatten((dequantize_per_tensor_default_4,), self._out_spec) 2025-09-09T15:05:08.3384230Z 2025-09-09T15:05:08.3384572Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:08.3385023Z onverted model fx: GraphModule( 2025-09-09T15:05:08.3385485Z (conv1): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:05:08.3386101Z (conv2): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:05:08.3386561Z ) 2025-09-09T15:05:08.3386679Z 2025-09-09T15:05:08.3386684Z 2025-09-09T15:05:08.3386689Z 2025-09-09T15:05:08.3386799Z def forward(self, x): 2025-09-09T15:05:08.3387552Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.014958452433347702, 14, -128, 127, torch.int8); x = None 2025-09-09T15:05:08.3389177Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.014958452433347702, 14, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:05:08.3390223Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:05:08.3391084Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.016144542023539543, -16, -128, 127, torch.int8); conv1 = None 2025-09-09T15:05:08.3392593Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016144542023539543, -16, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:05:08.3393754Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:05:08.3394626Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.01024628710001707, -11, -128, 127, torch.int8); conv2 = None 2025-09-09T15:05:08.3395888Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01024628710001707, -11, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:05:08.3396821Z return dequantize_per_tensor_default_2 2025-09-09T15:05:08.3397100Z 2025-09-09T15:05:08.3397377Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:08.3397742Z diff: tensor([[[0.], 2025-09-09T15:05:08.3397950Z [0.], 2025-09-09T15:05:08.3398142Z [0.]], 2025-09-09T15:05:08.3398258Z 2025-09-09T15:05:08.3398335Z [[0.], 2025-09-09T15:05:08.3398514Z [0.], 2025-09-09T15:05:08.3398699Z [0.]], 2025-09-09T15:05:08.3398819Z 2025-09-09T15:05:08.3398892Z [[0.], 2025-09-09T15:05:08.3399079Z [0.], 2025-09-09T15:05:08.3399262Z [0.]]]) 2025-09-09T15:05:08.3399677Z PASSED 2025-09-09T15:05:08.3400381Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_per_channel_weight_bias PASSED 2025-09-09T15:05:08.3401363Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_relu_fusion model pt2e: GraphModule( 2025-09-09T15:05:08.3401972Z (conv): Module() 2025-09-09T15:05:08.3402175Z (bn): Module() 2025-09-09T15:05:08.3402487Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:08.3403416Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:08.3404523Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:05:08.3405023Z ) 2025-09-09T15:05:08.3405304Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:08.3406285Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:05:08.3407554Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2845, -0.3289, -0.3229]), max_val=tensor([0.2989, 0.2870, 0.2939])) 2025-09-09T15:05:08.3408201Z ) 2025-09-09T15:05:08.3408473Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:08.3409414Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:08.3410502Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T15:05:08.3410975Z ) 2025-09-09T15:05:08.3411149Z ) 2025-09-09T15:05:08.3411247Z 2025-09-09T15:05:08.3411264Z 2025-09-09T15:05:08.3411268Z 2025-09-09T15:05:08.3411358Z def forward(self, x): 2025-09-09T15:05:08.3411635Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:05:08.3411974Z conv_weight = self.conv.weight 2025-09-09T15:05:08.3412244Z conv_bias = self.conv.bias 2025-09-09T15:05:08.3412506Z bn_weight = self.bn.weight 2025-09-09T15:05:08.3414383Z bn_bias = self.bn.bias 2025-09-09T15:05:08.3414648Z bn_running_mean = self.bn.running_mean 2025-09-09T15:05:08.3414953Z bn_running_var = self.bn.running_var 2025-09-09T15:05:08.3415281Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:05:08.3415796Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:05:08.3416380Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:05:08.3416906Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:05:08.3417296Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:05:08.3417707Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:05:08.3418146Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:05:08.3418641Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:05:08.3419207Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:05:08.3419811Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:05:08.3420812Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:05:08.3421710Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:05:08.3422236Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:05:08.3422807Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:05:08.3423349Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:05:08.3424212Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:05:08.3425060Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:05:08.3425576Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:05:08.3426123Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:05:08.3426506Z 2025-09-09T15:05:08.3426789Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:08.3427144Z model fx: GraphModule( 2025-09-09T15:05:08.3427468Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:08.3428393Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:08.3429480Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:05:08.3429993Z ) 2025-09-09T15:05:08.3430172Z (conv): ConvBnReLU1d( 2025-09-09T15:05:08.3430406Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:05:08.3430813Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:05:28.0177962Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:28.0179208Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:05:28.0180845Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2845, -0.3289, -0.3229]), max_val=tensor([0.2989, 0.2870, 0.2939])) 2025-09-09T15:05:28.0181654Z ) 2025-09-09T15:05:28.0181863Z ) 2025-09-09T15:05:28.0182490Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:28.0183666Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:28.0185144Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T15:05:28.0185728Z ) 2025-09-09T15:05:28.0185924Z ) 2025-09-09T15:05:28.0186049Z 2025-09-09T15:05:28.0186054Z 2025-09-09T15:05:28.0186059Z 2025-09-09T15:05:28.0186159Z def forward(self, x): 2025-09-09T15:05:28.0186585Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:05:28.0187235Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:05:28.0187914Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:05:28.0188447Z return activation_post_process_1 2025-09-09T15:05:28.0188765Z 2025-09-09T15:05:28.0189097Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:28.0189551Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:05:28.0189825Z [0., 0., 0.], 2025-09-09T15:05:28.0190121Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:05:28.0190489Z converted model pt2e: GraphModule( 2025-09-09T15:05:28.0190823Z (conv): Module() 2025-09-09T15:05:28.0191066Z (bn): Module() 2025-09-09T15:05:28.0191290Z ) 2025-09-09T15:05:28.0191426Z 2025-09-09T15:05:28.0199300Z 2025-09-09T15:05:28.0199305Z 2025-09-09T15:05:28.0199442Z def forward(self, x): 2025-09-09T15:05:28.0199799Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:05:28.0200211Z conv_bias = self.conv.bias 2025-09-09T15:05:28.0200585Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:05:28.0201458Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:05:28.0203028Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:05:28.0204335Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:05:28.0204922Z _scale_0 = self._scale_0 2025-09-09T15:05:28.0205235Z _zero_point_0 = self._zero_point_0 2025-09-09T15:05:28.0205596Z quantize_per_channel = self._frozen_param0 2025-09-09T15:05:28.0206723Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:05:28.0208511Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:05:28.0209504Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T15:05:28.0210284Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005544713232666254, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:05:28.0211553Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:05:28.0212568Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:05:28.0212972Z 2025-09-09T15:05:28.0213257Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:28.0213638Z onverted model fx: GraphModule( 2025-09-09T15:05:28.0213889Z (conv): ConvReLU1d( 2025-09-09T15:05:28.0214339Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:05:28.0214695Z (1): ReLU() 2025-09-09T15:05:28.0214879Z ) 2025-09-09T15:05:28.0215043Z ) 2025-09-09T15:05:28.0215136Z 2025-09-09T15:05:28.0215141Z 2025-09-09T15:05:28.0215145Z 2025-09-09T15:05:28.0215230Z def forward(self, x): 2025-09-09T15:05:28.0215918Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:05:28.0217145Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:05:28.0218153Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:05:28.0218999Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005544713232666254, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:05:28.0220280Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:05:28.0221160Z return dequantize_per_tensor_default_1 2025-09-09T15:05:28.0221437Z 2025-09-09T15:05:28.0221721Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:28.0222083Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:05:28.0222318Z [0., 0., 0.], 2025-09-09T15:05:28.0222526Z [0., 0., 0.]]]) 2025-09-09T15:05:28.0222753Z model pt2e: GraphModule( 2025-09-09T15:05:28.0222974Z (conv): Module() 2025-09-09T15:05:28.0223176Z (bn): Module() 2025-09-09T15:05:28.0223469Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:28.0224403Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:28.0225502Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:05:28.0225999Z ) 2025-09-09T15:05:28.0226279Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:28.0227209Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:05:28.0228315Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3289433717727661, max_val=0.29890719056129456) 2025-09-09T15:05:28.0228826Z ) 2025-09-09T15:05:28.0229094Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:28.0230023Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:28.0231065Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T15:05:28.0231528Z ) 2025-09-09T15:05:28.0231691Z ) 2025-09-09T15:05:28.0231798Z 2025-09-09T15:05:28.0231802Z 2025-09-09T15:05:28.0231806Z 2025-09-09T15:05:28.0231890Z def forward(self, x): 2025-09-09T15:05:28.0232168Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:05:28.0232495Z conv_weight = self.conv.weight 2025-09-09T15:05:28.0232765Z conv_bias = self.conv.bias 2025-09-09T15:05:28.0233009Z bn_weight = self.bn.weight 2025-09-09T15:05:28.0233254Z bn_bias = self.bn.bias 2025-09-09T15:05:28.0233504Z bn_running_mean = self.bn.running_mean 2025-09-09T15:05:28.0233805Z bn_running_var = self.bn.running_var 2025-09-09T15:05:28.0234273Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:05:28.0234700Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:05:28.0235285Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:05:28.0235883Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:05:28.0236331Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:05:28.0236732Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:05:28.0237170Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:05:28.0237666Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:05:28.0238218Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:05:28.0238826Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:05:28.0239794Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:05:28.0240671Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:05:28.0241206Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:05:28.0241767Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:05:28.0242313Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:05:28.0243169Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:05:28.0244046Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:05:28.0244584Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:05:28.0245121Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:05:28.0245511Z 2025-09-09T15:05:28.0245790Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:28.0246155Z model fx: GraphModule( 2025-09-09T15:05:28.0246470Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:28.0247407Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:53.6672216Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:05:53.6674134Z ) 2025-09-09T15:05:53.6674373Z (conv): ConvBnReLU1d( 2025-09-09T15:05:53.6674624Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:05:53.6675066Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:05:53.6675531Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:53.6676552Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:05:53.6677680Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3289433717727661, max_val=0.29890719056129456) 2025-09-09T15:05:53.6678192Z ) 2025-09-09T15:05:53.6678366Z ) 2025-09-09T15:05:53.6678639Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:53.6679577Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0055]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:53.6680869Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.413901925086975) 2025-09-09T15:05:53.6681328Z ) 2025-09-09T15:05:53.6681500Z ) 2025-09-09T15:05:53.6681595Z 2025-09-09T15:05:53.6681600Z 2025-09-09T15:05:53.6681603Z 2025-09-09T15:05:53.6681836Z def forward(self, x): 2025-09-09T15:05:53.6682218Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:05:53.6682755Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:05:53.6683293Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:05:53.6683716Z return activation_post_process_1 2025-09-09T15:05:53.6683969Z 2025-09-09T15:05:53.6684249Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:53.6684609Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:05:53.6684846Z [0., 0., 0.], 2025-09-09T15:05:53.6685091Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:05:53.6685389Z converted model pt2e: GraphModule( 2025-09-09T15:05:53.6685656Z (conv): Module() 2025-09-09T15:05:53.6685861Z (bn): Module() 2025-09-09T15:05:53.6686059Z ) 2025-09-09T15:05:53.6686161Z 2025-09-09T15:05:53.6686166Z 2025-09-09T15:05:53.6686170Z 2025-09-09T15:05:53.6686264Z def forward(self, x): 2025-09-09T15:05:53.6686546Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:05:53.6686875Z conv_bias = self.conv.bias 2025-09-09T15:05:53.6687180Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:05:53.6687887Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:05:53.6689104Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:05:53.6690198Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:05:53.6690687Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:05:53.6691477Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002590105403214693, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:05:53.6692724Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:05:53.6693560Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T15:05:53.6694337Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005544713232666254, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:05:53.6695611Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:05:53.6696616Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:05:53.6697030Z 2025-09-09T15:05:53.6697310Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:53.6697689Z onverted model fx: GraphModule( 2025-09-09T15:05:53.6697944Z (conv): ConvReLU1d( 2025-09-09T15:05:53.6698272Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:05:53.6698636Z (1): ReLU() 2025-09-09T15:05:53.6698822Z ) 2025-09-09T15:05:53.6698998Z ) 2025-09-09T15:05:53.6699094Z 2025-09-09T15:05:53.6699098Z 2025-09-09T15:05:53.6699102Z 2025-09-09T15:05:53.6699190Z def forward(self, x): 2025-09-09T15:05:53.6699801Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:05:53.6701229Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:05:53.6702310Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:05:53.6703174Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005544713232666254, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:05:53.6704451Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005544713232666254, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:05:53.6705342Z return dequantize_per_tensor_default_1 2025-09-09T15:05:53.6705619Z 2025-09-09T15:05:53.6705895Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:05:53.6706270Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:05:53.6706497Z [0., 0., 0.], 2025-09-09T15:05:53.6706702Z [0., 0., 0.]]]) 2025-09-09T15:05:53.6707130Z PASSED 2025-09-09T15:05:53.6707747Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_relu_fusion_cuda model pt2e: GraphModule( 2025-09-09T15:05:53.6708384Z (conv): Module() 2025-09-09T15:05:53.6708580Z (bn): Module() 2025-09-09T15:05:53.6708886Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:53.6710007Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:53.6711332Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:05:53.6711846Z ) 2025-09-09T15:05:53.6712123Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:53.6713299Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:05:53.6714847Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T15:05:53.6715579Z ) 2025-09-09T15:05:53.6715857Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:05:53.6717027Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:05:53.6718268Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123148918151855) 2025-09-09T15:05:53.6718731Z ) 2025-09-09T15:05:53.6718911Z ) 2025-09-09T15:05:53.6719006Z 2025-09-09T15:05:53.6719011Z 2025-09-09T15:05:53.6719015Z 2025-09-09T15:05:53.6719111Z def forward(self, x): 2025-09-09T15:05:53.6719389Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:05:53.6719727Z conv_weight = self.conv.weight 2025-09-09T15:05:53.6719995Z conv_bias = self.conv.bias 2025-09-09T15:05:53.6720249Z bn_weight = self.bn.weight 2025-09-09T15:05:53.6720496Z bn_bias = self.bn.bias 2025-09-09T15:05:53.6720756Z bn_running_mean = self.bn.running_mean 2025-09-09T15:05:53.6721050Z bn_running_var = self.bn.running_var 2025-09-09T15:05:53.6721481Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:05:53.6721926Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:05:53.6722511Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:05:53.6724618Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:05:53.6725016Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:05:53.6725430Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:05:53.6725870Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:05:53.6726370Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:05:53.6726937Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:05:53.6727563Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:05:53.6728551Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:05:53.6729430Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:05:53.6729971Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:05:53.6730543Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:05:53.6731101Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:05:53.6731977Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:05:53.6732819Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:06:11.5351397Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:06:11.5352031Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:06:11.5352413Z 2025-09-09T15:06:11.5352703Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:11.5353080Z model fx: GraphModule( 2025-09-09T15:06:11.5353407Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:11.5354524Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:11.5355803Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:06:11.5356382Z ) 2025-09-09T15:06:11.5356578Z (conv): ConvBnReLU1d( 2025-09-09T15:06:11.5356820Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:06:11.5357220Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:06:11.5357686Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:11.5358825Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022, 0.0020, 0.0022], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:06:11.5360372Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2799, -0.2557, -0.2618], device='cuda:0'), max_val=tensor([0.1970, 0.2308, 0.2775], device='cuda:0')) 2025-09-09T15:06:11.5361103Z ) 2025-09-09T15:06:11.5361272Z ) 2025-09-09T15:06:11.5361555Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:11.5363082Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:11.5364612Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123148918151855) 2025-09-09T15:06:11.5365090Z ) 2025-09-09T15:06:11.5365258Z ) 2025-09-09T15:06:11.5365361Z 2025-09-09T15:06:11.5365365Z 2025-09-09T15:06:11.5365369Z 2025-09-09T15:06:11.5365453Z def forward(self, x): 2025-09-09T15:06:11.5365805Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:06:11.5366336Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:06:11.5366884Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:06:11.5367314Z return activation_post_process_1 2025-09-09T15:06:11.5367576Z 2025-09-09T15:06:11.5367849Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:11.5368221Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:06:11.5368462Z [0., 0., 0.], 2025-09-09T15:06:11.5368738Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T15:06:11.5369077Z converted model pt2e: GraphModule( 2025-09-09T15:06:11.5369341Z (conv): Module() 2025-09-09T15:06:11.5369550Z (bn): Module() 2025-09-09T15:06:11.5369738Z ) 2025-09-09T15:06:11.5369840Z 2025-09-09T15:06:11.5369844Z 2025-09-09T15:06:11.5369848Z 2025-09-09T15:06:11.5369931Z def forward(self, x): 2025-09-09T15:06:11.5370212Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:06:11.5370549Z conv_bias = self.conv.bias 2025-09-09T15:06:11.5370851Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:06:11.5371556Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:06:11.5372795Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:06:11.5373837Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:06:11.5374304Z _scale_0 = self._scale_0 2025-09-09T15:06:11.5374564Z _zero_point_0 = self._zero_point_0 2025-09-09T15:06:11.5374866Z quantize_per_channel = self._frozen_param0 2025-09-09T15:06:11.5375755Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:06:11.5377098Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:06:11.5377957Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T15:06:11.5378744Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005538490135222673, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:06:11.5380018Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005538490135222673, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:06:11.5381034Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:06:11.5381452Z 2025-09-09T15:06:11.5381727Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:11.5382106Z onverted model fx: GraphModule( 2025-09-09T15:06:11.5382358Z (conv): ConvReLU1d( 2025-09-09T15:06:11.5382832Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:06:11.5383185Z (1): ReLU() 2025-09-09T15:06:11.5383377Z ) 2025-09-09T15:06:11.5383542Z ) 2025-09-09T15:06:11.5383644Z 2025-09-09T15:06:11.5383648Z 2025-09-09T15:06:11.5383652Z 2025-09-09T15:06:11.5383734Z def forward(self, x): 2025-09-09T15:06:11.5384418Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:06:11.5385638Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:06:11.5386639Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:06:11.5387486Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005538490135222673, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:06:11.5388768Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005538490135222673, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:06:11.5389655Z return dequantize_per_tensor_default_1 2025-09-09T15:06:11.5389924Z 2025-09-09T15:06:11.5390204Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:11.5390562Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:06:11.5390797Z [0., 0., 0.], 2025-09-09T15:06:11.5391053Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T15:06:11.5391336Z model pt2e: GraphModule( 2025-09-09T15:06:11.5391569Z (conv): Module() 2025-09-09T15:06:11.5391762Z (bn): Module() 2025-09-09T15:06:11.5392060Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:11.5393158Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:11.5394429Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:06:11.5394938Z ) 2025-09-09T15:06:11.5395209Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:11.5396387Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:06:11.5397657Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T15:06:11.5398174Z ) 2025-09-09T15:06:11.5398447Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:11.5399552Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:11.5400773Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123148918151855) 2025-09-09T15:06:11.5401234Z ) 2025-09-09T15:06:11.5401403Z ) 2025-09-09T15:06:11.5401500Z 2025-09-09T15:06:11.5401504Z 2025-09-09T15:06:11.5401508Z 2025-09-09T15:06:11.5401595Z def forward(self, x): 2025-09-09T15:06:11.5401880Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:06:11.5402217Z conv_weight = self.conv.weight 2025-09-09T15:06:11.5402489Z conv_bias = self.conv.bias 2025-09-09T15:06:11.5402837Z bn_weight = self.bn.weight 2025-09-09T15:06:11.5403079Z bn_bias = self.bn.bias 2025-09-09T15:06:11.5403340Z bn_running_mean = self.bn.running_mean 2025-09-09T15:06:11.5403633Z bn_running_var = self.bn.running_var 2025-09-09T15:06:11.5403965Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:06:11.5404519Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:06:11.5405106Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:06:11.5405633Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:06:11.5406022Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:06:11.5406434Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:06:11.5406872Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:06:11.5407372Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:06:11.5407936Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:06:11.5408540Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:06:37.2186426Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:06:37.2187575Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:06:37.2188234Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:06:37.2188943Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:06:37.2189622Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:06:37.2190711Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:06:37.2191778Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:06:37.2192425Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:06:37.2193108Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:06:37.2193587Z 2025-09-09T15:06:37.2193946Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:37.2194399Z model fx: GraphModule( 2025-09-09T15:06:37.2194784Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:37.2196264Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0104], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:37.2197942Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:06:37.2198576Z ) 2025-09-09T15:06:37.2198806Z (conv): ConvBnReLU1d( 2025-09-09T15:06:37.2199087Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:06:37.2199592Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:06:37.2200156Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:37.2201535Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0022], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:06:37.2203169Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.2799264192581177, max_val=0.27745386958122253) 2025-09-09T15:06:37.2204143Z ) 2025-09-09T15:06:37.2204352Z ) 2025-09-09T15:06:37.2204678Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:37.2206274Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0055], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:37.2207805Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.4123148918151855) 2025-09-09T15:06:37.2208382Z ) 2025-09-09T15:06:37.2208593Z ) 2025-09-09T15:06:37.2208710Z 2025-09-09T15:06:37.2208715Z 2025-09-09T15:06:37.2208720Z 2025-09-09T15:06:37.2208822Z def forward(self, x): 2025-09-09T15:06:37.2209250Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:06:37.2209909Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:06:37.2210595Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:06:37.2211126Z return activation_post_process_1 2025-09-09T15:06:37.2211442Z 2025-09-09T15:06:37.2211775Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:37.2212234Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:06:37.2212520Z [0., 0., 0.], 2025-09-09T15:06:37.2212826Z [0., 0., 0.]]], device='cuda:0', grad_fn=) 2025-09-09T15:06:37.2213228Z converted model pt2e: GraphModule( 2025-09-09T15:06:37.2213547Z (conv): Module() 2025-09-09T15:06:37.2213786Z (bn): Module() 2025-09-09T15:06:37.2214020Z ) 2025-09-09T15:06:37.2214138Z 2025-09-09T15:06:37.2214143Z 2025-09-09T15:06:37.2214148Z 2025-09-09T15:06:37.2214251Z def forward(self, x): 2025-09-09T15:06:37.2214597Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:06:37.2214992Z conv_bias = self.conv.bias 2025-09-09T15:06:37.2215360Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:06:37.2216237Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:06:37.2217814Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:06:37.2219131Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:06:37.2219731Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:06:37.2220731Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002204145072028041, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:06:37.2222334Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:06:37.2223474Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T15:06:37.2224308Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.005538490135222673, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:06:37.2225565Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.005538490135222673, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:06:37.2226572Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:06:37.2226989Z 2025-09-09T15:06:37.2227266Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:37.2227687Z onverted model fx: GraphModule( 2025-09-09T15:06:37.2227935Z (conv): ConvReLU1d( 2025-09-09T15:06:37.2228369Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:06:37.2228719Z (1): ReLU() 2025-09-09T15:06:37.2228909Z ) 2025-09-09T15:06:37.2229093Z ) 2025-09-09T15:06:37.2236360Z 2025-09-09T15:06:37.2236365Z 2025-09-09T15:06:37.2236369Z 2025-09-09T15:06:37.2236489Z def forward(self, x): 2025-09-09T15:06:37.2237232Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933931648731, 0, -128, 127, torch.int8); x = None 2025-09-09T15:06:37.2238513Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933931648731, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:06:37.2239516Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:06:37.2240361Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.005538490135222673, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:06:37.2241644Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.005538490135222673, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:06:37.2242531Z return dequantize_per_tensor_default_1 2025-09-09T15:06:37.2242809Z 2025-09-09T15:06:37.2243096Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:37.2243466Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:06:37.2243707Z [0., 0., 0.], 2025-09-09T15:06:37.2243929Z [0., 0., 0.]]], device='cuda:0') 2025-09-09T15:06:37.2244417Z PASSED 2025-09-09T15:06:37.2245044Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_relu_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T15:06:37.2245704Z (conv): Module() 2025-09-09T15:06:37.2245928Z (bn): Module() 2025-09-09T15:06:37.2246231Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:37.2247164Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:37.2248257Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:06:37.2248769Z ) 2025-09-09T15:06:37.2249044Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:37.2250018Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:06:37.2251291Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3276, -0.3045, -0.2418]), max_val=tensor([0.2760, 0.3298, 0.3101])) 2025-09-09T15:06:37.2251930Z ) 2025-09-09T15:06:37.2252207Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:37.2253140Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:37.2254188Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3778926134109497) 2025-09-09T15:06:37.2254654Z ) 2025-09-09T15:06:37.2254825Z ) 2025-09-09T15:06:37.2254923Z 2025-09-09T15:06:37.2254928Z 2025-09-09T15:06:37.2254932Z 2025-09-09T15:06:37.2255023Z def forward(self, x): 2025-09-09T15:06:37.2255307Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:06:37.2255648Z conv_weight = self.conv.weight 2025-09-09T15:06:37.2255915Z bn_weight = self.bn.weight 2025-09-09T15:06:37.2256276Z bn_bias = self.bn.bias 2025-09-09T15:06:37.2256530Z bn_running_mean = self.bn.running_mean 2025-09-09T15:06:37.2256831Z bn_running_var = self.bn.running_var 2025-09-09T15:06:37.2257163Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:06:37.2257593Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:06:37.2258258Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:06:57.0548901Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:06:57.0549444Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:06:57.0549949Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:06:57.0550486Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:06:57.0551098Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:06:57.0551799Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:06:57.0552875Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:06:57.0553898Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:06:57.0554562Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:06:57.0555665Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:06:57.0556834Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:06:57.0557480Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:06:57.0558151Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:06:57.0558643Z 2025-09-09T15:06:57.0558979Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:57.0559427Z model fx: GraphModule( 2025-09-09T15:06:57.0559812Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:57.0560991Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:57.0562386Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:06:57.0563021Z ) 2025-09-09T15:06:57.0563246Z (conv): ConvBnReLU1d( 2025-09-09T15:06:57.0563557Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:06:57.0564257Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:06:57.0564841Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:57.0566034Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026, 0.0026, 0.0025]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:06:57.0567673Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3276, -0.3045, -0.2418]), max_val=tensor([0.2760, 0.3298, 0.3101])) 2025-09-09T15:06:57.0568481Z ) 2025-09-09T15:06:57.0568694Z ) 2025-09-09T15:06:57.0569031Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:57.0570209Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:57.0571606Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3778926134109497) 2025-09-09T15:06:57.0572522Z ) 2025-09-09T15:06:57.0572741Z ) 2025-09-09T15:06:57.0572860Z 2025-09-09T15:06:57.0572866Z 2025-09-09T15:06:57.0572870Z 2025-09-09T15:06:57.0572984Z def forward(self, x): 2025-09-09T15:06:57.0573568Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:06:57.0574237Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:06:57.0574908Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:06:57.0575437Z return activation_post_process_1 2025-09-09T15:06:57.0575749Z 2025-09-09T15:06:57.0576086Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:57.0576537Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:06:57.0576825Z [0., 0., 0.], 2025-09-09T15:06:57.0577110Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:06:57.0577471Z converted model pt2e: GraphModule( 2025-09-09T15:06:57.0577798Z (conv): Module() 2025-09-09T15:06:57.0578040Z (bn): Module() 2025-09-09T15:06:57.0578276Z ) 2025-09-09T15:06:57.0578395Z 2025-09-09T15:06:57.0578400Z 2025-09-09T15:06:57.0578405Z 2025-09-09T15:06:57.0578511Z def forward(self, x): 2025-09-09T15:06:57.0578862Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:06:57.0579315Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:06:57.0580178Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:06:57.0581701Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:06:57.0582972Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:06:57.0583555Z _scale_0 = self._scale_0 2025-09-09T15:06:57.0583873Z _zero_point_0 = self._zero_point_0 2025-09-09T15:06:57.0584234Z quantize_per_channel = self._frozen_param0 2025-09-09T15:06:57.0585329Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:06:57.0586407Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:06:57.0587441Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T15:06:57.0588587Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T15:06:57.0589630Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0054035005159676075, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:06:57.0590935Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0054035005159676075, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:06:57.0591954Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:06:57.0592372Z 2025-09-09T15:06:57.0592660Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:57.0593033Z onverted model fx: GraphModule( 2025-09-09T15:06:57.0593292Z (conv): ConvReLU1d( 2025-09-09T15:06:57.0593620Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:06:57.0593990Z (1): ReLU() 2025-09-09T15:06:57.0594182Z ) 2025-09-09T15:06:57.0594360Z ) 2025-09-09T15:06:57.0594457Z 2025-09-09T15:06:57.0594462Z 2025-09-09T15:06:57.0594465Z 2025-09-09T15:06:57.0594552Z def forward(self, x): 2025-09-09T15:06:57.0595167Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:06:57.0596623Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:06:57.0597626Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:06:57.0598487Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0054035005159676075, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:06:57.0599778Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0054035005159676075, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:06:57.0600713Z return dequantize_per_tensor_default_1 2025-09-09T15:06:57.0600999Z 2025-09-09T15:06:57.0601280Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:06:57.0601650Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:06:57.0601880Z [0., 0., 0.], 2025-09-09T15:06:57.0602097Z [0., 0., 0.]]]) 2025-09-09T15:06:57.0602330Z model pt2e: GraphModule( 2025-09-09T15:06:57.0602566Z (conv): Module() 2025-09-09T15:06:57.0602775Z (bn): Module() 2025-09-09T15:06:57.0603077Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:57.0604014Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:57.0605099Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:06:57.0605611Z ) 2025-09-09T15:06:57.0605903Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:57.0606848Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:06:57.0607964Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.32764676213264465, max_val=0.3298276662826538) 2025-09-09T15:06:57.0608475Z ) 2025-09-09T15:06:57.0608761Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:06:57.0609711Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:06:57.0610803Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3749419450759888) 2025-09-09T15:06:57.0611283Z ) 2025-09-09T15:06:57.0611464Z ) 2025-09-09T15:06:57.0611565Z 2025-09-09T15:06:57.0611569Z 2025-09-09T15:06:57.0611573Z 2025-09-09T15:06:57.0611668Z def forward(self, x): 2025-09-09T15:06:57.0611952Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:06:57.0612293Z conv_weight = self.conv.weight 2025-09-09T15:06:57.0612567Z bn_weight = self.bn.weight 2025-09-09T15:06:57.0612828Z bn_bias = self.bn.bias 2025-09-09T15:06:57.0613090Z bn_running_mean = self.bn.running_mean 2025-09-09T15:06:57.0613397Z bn_running_var = self.bn.running_var 2025-09-09T15:06:57.0613727Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:06:57.0614168Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:06:57.0614761Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:07:15.3660438Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:07:15.3661347Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:07:15.3662503Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:07:15.3662973Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:07:15.3663465Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:07:15.3664377Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:07:15.3665209Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:07:15.3666028Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:07:15.3666559Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:07:15.3667436Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:07:15.3668278Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:07:15.3668796Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:07:15.3669333Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:07:15.3669720Z 2025-09-09T15:07:15.3670004Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:15.3670365Z model fx: GraphModule( 2025-09-09T15:07:15.3670686Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:15.3671626Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:15.3672998Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:15.3681503Z ) 2025-09-09T15:07:15.3681875Z (conv): ConvBnReLU1d( 2025-09-09T15:07:15.3682271Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:07:15.3682860Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:07:15.3683407Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:15.3684331Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:07:15.3685458Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.32764676213264465, max_val=0.3298276662826538) 2025-09-09T15:07:15.3685974Z ) 2025-09-09T15:07:15.3686149Z ) 2025-09-09T15:07:15.3686434Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:15.3687390Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0054]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:15.3688475Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.3749419450759888) 2025-09-09T15:07:15.3688940Z ) 2025-09-09T15:07:15.3689113Z ) 2025-09-09T15:07:15.3689210Z 2025-09-09T15:07:15.3689214Z 2025-09-09T15:07:15.3689218Z 2025-09-09T15:07:15.3689327Z def forward(self, x): 2025-09-09T15:07:15.3689674Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:15.3690209Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:07:15.3690754Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:07:15.3691174Z return activation_post_process_1 2025-09-09T15:07:15.3691623Z 2025-09-09T15:07:15.3691897Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:15.3692268Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:15.3692502Z [0., 0., 0.], 2025-09-09T15:07:15.3692737Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:07:15.3693035Z converted model pt2e: GraphModule( 2025-09-09T15:07:15.3693416Z (conv): Module() 2025-09-09T15:07:15.3693621Z (bn): Module() 2025-09-09T15:07:15.3693811Z ) 2025-09-09T15:07:15.3693907Z 2025-09-09T15:07:15.3693911Z 2025-09-09T15:07:15.3693915Z 2025-09-09T15:07:15.3694007Z def forward(self, x): 2025-09-09T15:07:15.3694286Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:15.3694661Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:07:15.3695359Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:15.3696587Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:15.3697633Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:07:15.3698126Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:07:15.3698911Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002597068203613162, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:07:15.3699706Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:07:15.3700533Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T15:07:15.3701620Z relu = torch.ops.aten.relu.default(conv1d_2); conv1d_2 = None 2025-09-09T15:07:15.3702561Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0053919292986392975, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:07:15.3704167Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0053919292986392975, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:07:15.3705186Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:07:15.3705596Z 2025-09-09T15:07:15.3705885Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:15.3706253Z onverted model fx: GraphModule( 2025-09-09T15:07:15.3706517Z (conv): ConvReLU1d( 2025-09-09T15:07:15.3706840Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:07:15.3707202Z (1): ReLU() 2025-09-09T15:07:15.3707394Z ) 2025-09-09T15:07:15.3707569Z ) 2025-09-09T15:07:15.3707672Z 2025-09-09T15:07:15.3707677Z 2025-09-09T15:07:15.3707681Z 2025-09-09T15:07:15.3707771Z def forward(self, x): 2025-09-09T15:07:15.3708375Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:15.3709610Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:15.3710620Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:07:15.3711477Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0053919292986392975, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:07:15.3712814Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0053919292986392975, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:15.3713784Z return dequantize_per_tensor_default_1 2025-09-09T15:07:15.3714060Z 2025-09-09T15:07:15.3714342Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:15.3714703Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:15.3715009Z [0., 0., 0.], 2025-09-09T15:07:15.3715215Z [0., 0., 0.]]]) 2025-09-09T15:07:15.3715649Z PASSED 2025-09-09T15:07:15.3716308Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_no_bias model pt2e: GraphModule( 2025-09-09T15:07:15.3716907Z (conv): Module() 2025-09-09T15:07:15.3717210Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:15.3718201Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:07:15.3719492Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T15:07:15.3720129Z ) 2025-09-09T15:07:15.3720412Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:15.3721337Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:15.3722426Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:15.3722931Z ) 2025-09-09T15:07:15.3723203Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:15.3724141Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0038]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:15.3725197Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9578298330307007) 2025-09-09T15:07:15.3725659Z ) 2025-09-09T15:07:15.3725829Z ) 2025-09-09T15:07:15.3725932Z 2025-09-09T15:07:15.3725936Z 2025-09-09T15:07:15.3725940Z 2025-09-09T15:07:15.3726023Z def forward(self, x): 2025-09-09T15:07:15.3726306Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:15.3726635Z conv_weight = self.conv.weight 2025-09-09T15:07:15.3727093Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:07:15.3727676Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:15.3728472Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:07:15.3729237Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T15:07:16.9058679Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:07:16.9059416Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:07:16.9059924Z 2025-09-09T15:07:16.9060267Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:16.9060727Z model fx: GraphModule( 2025-09-09T15:07:16.9061122Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9062319Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:16.9064037Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:16.9064935Z ) 2025-09-09T15:07:16.9065155Z (conv): ConvReLU1d( 2025-09-09T15:07:16.9065452Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:07:16.9065887Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9067230Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024, 0.0022, 0.0021]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:07:16.9068857Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3119, -0.2799, -0.2618]), max_val=tensor([0.1970, 0.1855, 0.2308])) 2025-09-09T15:07:16.9069673Z ) 2025-09-09T15:07:16.9069876Z ) 2025-09-09T15:07:16.9070209Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9071388Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0038]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:16.9072721Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9578298330307007) 2025-09-09T15:07:16.9073356Z ) 2025-09-09T15:07:16.9073553Z ) 2025-09-09T15:07:16.9073677Z 2025-09-09T15:07:16.9073689Z 2025-09-09T15:07:16.9073694Z 2025-09-09T15:07:16.9073795Z def forward(self, x): 2025-09-09T15:07:16.9074220Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:16.9074866Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:07:16.9075542Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:07:16.9076060Z return activation_post_process_1 2025-09-09T15:07:16.9076473Z 2025-09-09T15:07:16.9076802Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:16.9077257Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:16.9077535Z [0., 0., 0.], 2025-09-09T15:07:16.9077819Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:07:16.9078188Z converted model pt2e: GraphModule( 2025-09-09T15:07:16.9078501Z (conv): Module() 2025-09-09T15:07:16.9078740Z ) 2025-09-09T15:07:16.9078856Z 2025-09-09T15:07:16.9078861Z 2025-09-09T15:07:16.9078872Z 2025-09-09T15:07:16.9078976Z def forward(self, x): 2025-09-09T15:07:16.9079316Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:16.9079710Z _scale_0 = self._scale_0 2025-09-09T15:07:16.9080016Z _zero_point_0 = self._zero_point_0 2025-09-09T15:07:16.9080400Z quantize_per_channel_default = self._frozen_param0 2025-09-09T15:07:16.9081660Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T15:07:16.9083343Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:16.9084895Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:16.9086556Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T15:07:16.9087564Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T15:07:16.9088500Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0037561955396085978, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:07:16.9090095Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0037561955396085978, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:16.9091403Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:07:16.9091815Z 2025-09-09T15:07:16.9092191Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:16.9092565Z onverted model fx: GraphModule( 2025-09-09T15:07:16.9092820Z (conv): ConvReLU1d( 2025-09-09T15:07:16.9093173Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T15:07:16.9093569Z (1): ReLU() 2025-09-09T15:07:16.9093752Z ) 2025-09-09T15:07:16.9093920Z ) 2025-09-09T15:07:16.9094014Z 2025-09-09T15:07:16.9094018Z 2025-09-09T15:07:16.9094022Z 2025-09-09T15:07:16.9094118Z def forward(self, x): 2025-09-09T15:07:16.9094721Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:16.9095946Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:16.9096950Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:07:16.9097801Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0037561955396085978, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:07:16.9099078Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0037561955396085978, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:16.9099952Z return dequantize_per_tensor_default_1 2025-09-09T15:07:16.9108402Z 2025-09-09T15:07:16.9108750Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:16.9109191Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:16.9109443Z [0., 0., 0.], 2025-09-09T15:07:16.9109672Z [0., 0., 0.]]]) 2025-09-09T15:07:16.9109918Z model pt2e: GraphModule( 2025-09-09T15:07:16.9110155Z (conv): Module() 2025-09-09T15:07:16.9110467Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9111411Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:07:16.9112522Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3119288384914398, max_val=0.23078612983226776) 2025-09-09T15:07:16.9113034Z ) 2025-09-09T15:07:16.9113321Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9114247Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:16.9115325Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:16.9115831Z ) 2025-09-09T15:07:16.9116108Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9117099Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0037]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:16.9118150Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9556508660316467) 2025-09-09T15:07:16.9118614Z ) 2025-09-09T15:07:16.9118790Z ) 2025-09-09T15:07:16.9118892Z 2025-09-09T15:07:16.9118896Z 2025-09-09T15:07:16.9118900Z 2025-09-09T15:07:16.9118984Z def forward(self, x): 2025-09-09T15:07:16.9119393Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:16.9119722Z conv_weight = self.conv.weight 2025-09-09T15:07:16.9120180Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:07:16.9120837Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:16.9121640Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:07:16.9122393Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T15:07:16.9122880Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:07:16.9123461Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:07:16.9123842Z 2025-09-09T15:07:16.9124122Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:16.9124487Z model fx: GraphModule( 2025-09-09T15:07:16.9124812Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9125744Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:16.9126819Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:16.9127325Z ) 2025-09-09T15:07:16.9127500Z (conv): ConvReLU1d( 2025-09-09T15:07:16.9127754Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:07:16.9128111Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:16.9129022Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0024]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:07:16.9130131Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3119288384914398, max_val=0.23078612983226776) 2025-09-09T15:07:16.9130637Z ) 2025-09-09T15:07:16.9130808Z ) 2025-09-09T15:07:16.9131073Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6354905Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0037]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:18.6356429Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.9556508660316467) 2025-09-09T15:07:18.6357022Z ) 2025-09-09T15:07:18.6357222Z ) 2025-09-09T15:07:18.6357340Z 2025-09-09T15:07:18.6357354Z 2025-09-09T15:07:18.6357359Z 2025-09-09T15:07:18.6357462Z def forward(self, x): 2025-09-09T15:07:18.6357890Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:18.6358561Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:07:18.6359242Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:07:18.6359765Z return activation_post_process_1 2025-09-09T15:07:18.6360088Z 2025-09-09T15:07:18.6360429Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:18.6360882Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:18.6361160Z [0., 0., 0.], 2025-09-09T15:07:18.6361447Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:07:18.6361814Z converted model pt2e: GraphModule( 2025-09-09T15:07:18.6362131Z (conv): Module() 2025-09-09T15:07:18.6362369Z ) 2025-09-09T15:07:18.6362485Z 2025-09-09T15:07:18.6362491Z 2025-09-09T15:07:18.6362495Z 2025-09-09T15:07:18.6362596Z def forward(self, x): 2025-09-09T15:07:18.6362935Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:18.6363618Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T15:07:18.6365012Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0024561325553804636, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:18.6368202Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:18.6369775Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:18.6371442Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:07:18.6372458Z relu = torch.ops.aten.relu.default(conv1d); conv1d = None 2025-09-09T15:07:18.6373607Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.003747650422155857, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:07:18.6375253Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003747650422155857, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:07:18.6376534Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:07:18.6377045Z 2025-09-09T15:07:18.6377376Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:18.6377843Z onverted model fx: GraphModule( 2025-09-09T15:07:18.6378155Z (conv): ConvReLU1d( 2025-09-09T15:07:18.6378588Z (0): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T15:07:18.6379080Z (1): ReLU() 2025-09-09T15:07:18.6379308Z ) 2025-09-09T15:07:18.6379519Z ) 2025-09-09T15:07:18.6379633Z 2025-09-09T15:07:18.6379638Z 2025-09-09T15:07:18.6379643Z 2025-09-09T15:07:18.6379744Z def forward(self, x): 2025-09-09T15:07:18.6380495Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:18.6382040Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:18.6383316Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:07:18.6384392Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.003747650422155857, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:07:18.6386025Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003747650422155857, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:18.6387149Z return dequantize_per_tensor_default_1 2025-09-09T15:07:18.6387479Z 2025-09-09T15:07:18.6387807Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:18.6388267Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:18.6388543Z [0., 0., 0.], 2025-09-09T15:07:18.6388797Z [0., 0., 0.]]]) 2025-09-09T15:07:18.6389071Z model pt2e: GraphModule( 2025-09-09T15:07:18.6389356Z (conv): Module() 2025-09-09T15:07:18.6389729Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6390969Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0025, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:07:18.6392608Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2203, -0.3233, -0.3086]), max_val=tensor([0.2796, 0.3026, 0.2405])) 2025-09-09T15:07:18.6393556Z ) 2025-09-09T15:07:18.6393893Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6395144Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:18.6396613Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:18.6397254Z ) 2025-09-09T15:07:18.6397578Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6398759Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-25], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:18.6400154Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.8935509324073792, max_val=1.3209781646728516) 2025-09-09T15:07:18.6400792Z ) 2025-09-09T15:07:18.6400996Z ) 2025-09-09T15:07:18.6401112Z 2025-09-09T15:07:18.6401117Z 2025-09-09T15:07:18.6401122Z 2025-09-09T15:07:18.6401228Z def forward(self, x): 2025-09-09T15:07:18.6401569Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:18.6401985Z conv_weight = self.conv.weight 2025-09-09T15:07:18.6402542Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:07:18.6403269Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:18.6404273Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:07:18.6405317Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T15:07:18.6406001Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:07:18.6406477Z 2025-09-09T15:07:18.6406817Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:18.6407252Z model fx: GraphModule( 2025-09-09T15:07:18.6407643Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6408806Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:18.6410197Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:18.6410834Z ) 2025-09-09T15:07:18.6411045Z (conv): Conv1d( 2025-09-09T15:07:18.6411329Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:07:18.6411740Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6412685Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022, 0.0025, 0.0024]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:07:18.6413958Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2203, -0.3233, -0.3086]), max_val=tensor([0.2796, 0.3026, 0.2405])) 2025-09-09T15:07:18.6414601Z ) 2025-09-09T15:07:18.6414778Z ) 2025-09-09T15:07:18.6415046Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:18.6415975Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-25], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:18.6417059Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.8935509324073792, max_val=1.3209781646728516) 2025-09-09T15:07:18.6417664Z ) 2025-09-09T15:07:18.6417827Z ) 2025-09-09T15:07:18.6417920Z 2025-09-09T15:07:18.6417924Z 2025-09-09T15:07:18.6417928Z 2025-09-09T15:07:18.6418007Z def forward(self, x): 2025-09-09T15:07:18.6418433Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:18.6418955Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:07:18.6419498Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:07:18.6419913Z return activation_post_process_1 2025-09-09T15:07:18.6420171Z 2025-09-09T15:07:18.6420442Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:18.6420801Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:18.6421034Z [0., 0., 0.], 2025-09-09T15:07:18.6421270Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:07:18.6421576Z converted model pt2e: GraphModule( 2025-09-09T15:07:18.6421829Z (conv): Module() 2025-09-09T15:07:18.6422031Z ) 2025-09-09T15:07:18.6422125Z 2025-09-09T15:07:18.6422129Z 2025-09-09T15:07:18.6422133Z 2025-09-09T15:07:18.6422219Z def forward(self, x): 2025-09-09T15:07:18.6422497Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:18.6422831Z _scale_0 = self._scale_0 2025-09-09T15:07:18.6423077Z _zero_point_0 = self._zero_point_0 2025-09-09T15:07:18.6423399Z quantize_per_channel_default = self._frozen_param0 2025-09-09T15:07:18.6424430Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T15:07:20.1589746Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:20.1591558Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:20.1593336Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T15:07:20.1594593Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.008684427477419376, -25, -128, 127, torch.int8); conv1d = None 2025-09-09T15:07:20.1595873Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008684427477419376, -25, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:20.1596976Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:07:20.1597396Z 2025-09-09T15:07:20.1597680Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:20.1598064Z onverted model fx: GraphModule( 2025-09-09T15:07:20.1598468Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T15:07:20.1598882Z ) 2025-09-09T15:07:20.1598982Z 2025-09-09T15:07:20.1598991Z 2025-09-09T15:07:20.1598995Z 2025-09-09T15:07:20.1599092Z def forward(self, x): 2025-09-09T15:07:20.1599705Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:20.1600929Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:20.1601923Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:07:20.1603077Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.008684427477419376, -25, -128, 127, torch.int8); conv = None 2025-09-09T15:07:20.1604483Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008684427477419376, -25, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:20.1605352Z return dequantize_per_tensor_default_1 2025-09-09T15:07:20.1605627Z 2025-09-09T15:07:20.1605904Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:20.1606274Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:20.1606503Z [0., 0., 0.], 2025-09-09T15:07:20.1606715Z [0., 0., 0.]]]) 2025-09-09T15:07:20.1606943Z model pt2e: GraphModule( 2025-09-09T15:07:20.1607165Z (conv): Module() 2025-09-09T15:07:20.1607476Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:20.1608412Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:07:20.1609537Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232726454734802, max_val=0.30256539583206177) 2025-09-09T15:07:20.1610047Z ) 2025-09-09T15:07:20.1610328Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:20.1611257Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:20.1612340Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:20.1612836Z ) 2025-09-09T15:07:20.1613118Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:20.1614042Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-26], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:20.1615168Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.887858510017395, max_val=1.3209781646728516) 2025-09-09T15:07:20.1615683Z ) 2025-09-09T15:07:20.1615847Z ) 2025-09-09T15:07:20.1615943Z 2025-09-09T15:07:20.1615953Z 2025-09-09T15:07:20.1615958Z 2025-09-09T15:07:20.1616040Z def forward(self, x): 2025-09-09T15:07:20.1616319Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:20.1616659Z conv_weight = self.conv.weight 2025-09-09T15:07:20.1617119Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:07:20.1617695Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:20.1618507Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:07:20.1619324Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T15:07:20.1619884Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:07:20.1620268Z 2025-09-09T15:07:20.1620551Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:20.1620919Z model fx: GraphModule( 2025-09-09T15:07:20.1621232Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:20.1622163Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:20.1623240Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:07:20.1623833Z ) 2025-09-09T15:07:20.1624011Z (conv): Conv1d( 2025-09-09T15:07:20.1624243Z 3, 3, kernel_size=(3,), stride=(1,), bias=False 2025-09-09T15:07:20.1624597Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:20.1625579Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:07:20.1626688Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3232726454734802, max_val=0.30256539583206177) 2025-09-09T15:07:20.1627192Z ) 2025-09-09T15:07:20.1627366Z ) 2025-09-09T15:07:20.1627644Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:07:20.1628567Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0087]), zero_point=tensor([-26], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:07:20.1629656Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.887858510017395, max_val=1.3209781646728516) 2025-09-09T15:07:20.1630152Z ) 2025-09-09T15:07:20.1630320Z ) 2025-09-09T15:07:20.1630417Z 2025-09-09T15:07:20.1630426Z 2025-09-09T15:07:20.1630430Z 2025-09-09T15:07:20.1630525Z def forward(self, x): 2025-09-09T15:07:20.1630866Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:07:20.1631397Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:07:20.1631939Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:07:20.1632362Z return activation_post_process_1 2025-09-09T15:07:20.1632619Z 2025-09-09T15:07:20.1632898Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:20.1633270Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:07:20.1633504Z [0., 0., 0.], 2025-09-09T15:07:20.1633740Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:07:20.1634036Z converted model pt2e: GraphModule( 2025-09-09T15:07:20.1634297Z (conv): Module() 2025-09-09T15:07:20.1634485Z ) 2025-09-09T15:07:20.1634588Z 2025-09-09T15:07:20.1634591Z 2025-09-09T15:07:20.1634602Z 2025-09-09T15:07:20.1634685Z def forward(self, x): 2025-09-09T15:07:20.1634960Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:07:20.1635335Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T15:07:20.1636291Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.002545453840866685, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:07:20.1637510Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:20.1638745Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:07:20.1640073Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:07:20.1641236Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.008662103675305843, -26, -128, 127, torch.int8); conv1d = None 2025-09-09T15:07:20.1642513Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.008662103675305843, -26, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:07:20.1643518Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:07:20.1644016Z 2025-09-09T15:07:20.1644299Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:07:20.1644670Z onverted model fx: GraphModule( 2025-09-09T15:07:20.1645074Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,), bias=False) 2025-09-09T15:07:20.1645472Z ) 2025-09-09T15:07:20.1645653Z 2025-09-09T15:07:20.1645657Z 2025-09-09T15:07:20.1645661Z 2025-09-09T15:07:20.1645745Z def forward(self, x): 2025-09-09T15:07:20.1646353Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:07:20.1647564Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:08:40.2008054Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:08:40.2009004Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.008662103675305843, -26, -128, 127, torch.int8); conv = None 2025-09-09T15:08:40.2010287Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008662103675305843, -26, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:08:40.2011160Z return dequantize_per_tensor_default_1 2025-09-09T15:08:40.2011437Z 2025-09-09T15:08:40.2011718Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:40.2012091Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:08:40.2012329Z [0., 0., 0.], 2025-09-09T15:08:40.2012532Z [0., 0., 0.]]]) 2025-09-09T15:08:40.2012972Z PASSED 2025-09-09T15:08:40.2013626Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_transpose_bn PASSED 2025-09-09T15:08:40.2014650Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_transpose_bn_relu PASSED 2025-09-09T15:08:40.2015587Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_inplace_add_relu model pt2e: GraphModule( 2025-09-09T15:08:40.2016194Z (conv): Module() 2025-09-09T15:08:40.2016506Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2017508Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:08:40.2018728Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2457]), max_val=tensor([-0.2457])) 2025-09-09T15:08:40.2019291Z ) 2025-09-09T15:08:40.2019567Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2020496Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:40.2021575Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T15:08:40.2022082Z ) 2025-09-09T15:08:40.2022360Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2023276Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:40.2024357Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T15:08:40.2024861Z ) 2025-09-09T15:08:40.2025152Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2026413Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:40.2027649Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T15:08:40.2028121Z ) 2025-09-09T15:08:40.2028287Z ) 2025-09-09T15:08:40.2028393Z 2025-09-09T15:08:40.2028397Z 2025-09-09T15:08:40.2028401Z 2025-09-09T15:08:40.2028483Z def forward(self, x): 2025-09-09T15:08:40.2028771Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:08:40.2029102Z conv_weight = self.conv.weight 2025-09-09T15:08:40.2029559Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:08:40.2030018Z conv_bias = self.conv.bias 2025-09-09T15:08:40.2030391Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:08:40.2031171Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T15:08:40.2031967Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T15:08:40.2032764Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T15:08:40.2033460Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:08:40.2033932Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T15:08:40.2034466Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:08:40.2034849Z 2025-09-09T15:08:40.2035125Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:40.2035486Z model fx: GraphModule( 2025-09-09T15:08:40.2035811Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2036804Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:40.2037895Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T15:08:40.2038396Z ) 2025-09-09T15:08:40.2038571Z (conv): Conv1d( 2025-09-09T15:08:40.2038789Z 1, 1, kernel_size=(1,), stride=(1,) 2025-09-09T15:08:40.2039119Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2040033Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:08:40.2041194Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2457]), max_val=tensor([-0.2457])) 2025-09-09T15:08:40.2041757Z ) 2025-09-09T15:08:40.2041926Z ) 2025-09-09T15:08:40.2042203Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2043143Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:40.2044233Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T15:08:40.2044749Z ) 2025-09-09T15:08:40.2044934Z (relu): ReLU(inplace=True) 2025-09-09T15:08:40.2045276Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:40.2046215Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:40.2047405Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T15:08:40.2047875Z ) 2025-09-09T15:08:40.2048044Z ) 2025-09-09T15:08:40.2048147Z 2025-09-09T15:08:40.2048229Z 2025-09-09T15:08:40.2048233Z 2025-09-09T15:08:40.2048318Z def forward(self, x): 2025-09-09T15:08:40.2048672Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:08:40.2049100Z conv = self.conv(activation_post_process_0) 2025-09-09T15:08:40.2049540Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:08:40.2050226Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T15:08:40.2050785Z relu = self.relu(add); add = None 2025-09-09T15:08:40.2051195Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:08:40.2051626Z return activation_post_process_2 2025-09-09T15:08:40.2051890Z 2025-09-09T15:08:40.2052160Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:40.2052577Z diff: tensor([[[0., 0., 0.]]], grad_fn=) 2025-09-09T15:08:40.2052908Z converted model pt2e: GraphModule( 2025-09-09T15:08:40.2053177Z (conv): Module() 2025-09-09T15:08:40.2053369Z ) 2025-09-09T15:08:40.2053472Z 2025-09-09T15:08:40.2053476Z 2025-09-09T15:08:40.2053480Z 2025-09-09T15:08:40.2053564Z def forward(self, x): 2025-09-09T15:08:40.2053836Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:08:40.2054163Z _scale_0 = self._scale_0 2025-09-09T15:08:40.2054423Z _zero_point_0 = self._zero_point_0 2025-09-09T15:08:40.2054738Z quantize_per_channel_default = self._frozen_param0 2025-09-09T15:08:40.2055731Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T15:08:40.2056695Z conv_bias = self.conv.bias 2025-09-09T15:08:40.2057370Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T15:08:40.2058488Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8) 2025-09-09T15:08:40.2059794Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:08:40.2061196Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_3, dequantize_per_channel_default, conv_bias); dequantize_per_tensor_default_3 = dequantize_per_channel_default = conv_bias = None 2025-09-09T15:08:40.2062451Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.0021822303533554077, 46, -128, 127, torch.int8); conv1d = None 2025-09-09T15:08:40.2063914Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:08:40.2065226Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_1, dequantize_per_tensor_default_4); dequantize_per_tensor_default_1 = dequantize_per_tensor_default_4 = None 2025-09-09T15:08:40.2066011Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:08:41.8403468Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.0015069034416228533, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T15:08:41.8405439Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:08:41.8406600Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:08:41.8407017Z 2025-09-09T15:08:41.8407314Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:41.8407698Z onverted model fx: GraphModule( 2025-09-09T15:08:41.8408085Z (conv): QuantizedConv1d(Reference)(1, 1, kernel_size=(1,), stride=(1,)) 2025-09-09T15:08:41.8408488Z (relu): ReLU(inplace=True) 2025-09-09T15:08:41.8408725Z ) 2025-09-09T15:08:41.8408828Z 2025-09-09T15:08:41.8408833Z 2025-09-09T15:08:41.8408836Z 2025-09-09T15:08:41.8408934Z def forward(self, x): 2025-09-09T15:08:41.8409555Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T15:08:41.8410796Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:08:41.8411692Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T15:08:41.8412418Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0021822303533554077, 46, -128, 127, torch.int8); conv = None 2025-09-09T15:08:41.8413691Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:08:41.8414892Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:08:41.8415530Z relu = self.relu(add); add = None 2025-09-09T15:08:41.8416235Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0015069034416228533, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:08:41.8417517Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:08:41.8418456Z return dequantize_per_tensor_default_2 2025-09-09T15:08:41.8418732Z 2025-09-09T15:08:41.8419019Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:41.8419388Z diff: tensor([[[0., 0., 0.]]]) 2025-09-09T15:08:41.8419648Z model pt2e: GraphModule( 2025-09-09T15:08:41.8419886Z (conv): Module() 2025-09-09T15:08:41.8420195Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8421157Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:08:41.8422295Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.24565386772155762, max_val=-0.24565386772155762) 2025-09-09T15:08:41.8422829Z ) 2025-09-09T15:08:41.8423110Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8424044Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:41.8425145Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T15:08:41.8425651Z ) 2025-09-09T15:08:41.8425937Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8426953Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:41.8428161Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T15:08:41.8428697Z ) 2025-09-09T15:08:41.8428975Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8429921Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:41.8430971Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T15:08:41.8431437Z ) 2025-09-09T15:08:41.8431617Z ) 2025-09-09T15:08:41.8431719Z 2025-09-09T15:08:41.8431729Z 2025-09-09T15:08:41.8431733Z 2025-09-09T15:08:41.8431819Z def forward(self, x): 2025-09-09T15:08:41.8432106Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:08:41.8432441Z conv_weight = self.conv.weight 2025-09-09T15:08:41.8432908Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:08:41.8433390Z conv_bias = self.conv.bias 2025-09-09T15:08:41.8433760Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:08:41.8434542Z conv1d = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T15:08:41.8435342Z activation_post_process_2 = self.activation_post_process_2(conv1d); conv1d = None 2025-09-09T15:08:41.8436146Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T15:08:41.8436959Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:08:41.8437433Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T15:08:41.8437986Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:08:41.8438372Z 2025-09-09T15:08:41.8438664Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:41.8439024Z model fx: GraphModule( 2025-09-09T15:08:41.8439352Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8440288Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0089]), zero_point=tensor([41], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:41.8441374Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.498010516166687, max_val=0.7672448754310608) 2025-09-09T15:08:41.8441885Z ) 2025-09-09T15:08:41.8442072Z (conv): Conv1d( 2025-09-09T15:08:41.8442303Z 1, 1, kernel_size=(1,), stride=(1,) 2025-09-09T15:08:41.8442642Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8443571Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0010]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:08:41.8444697Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.24565386772155762, max_val=-0.24565386772155762) 2025-09-09T15:08:41.8445220Z ) 2025-09-09T15:08:41.8445398Z ) 2025-09-09T15:08:41.8445680Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8446617Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0022]), zero_point=tensor([46], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:41.8447862Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3805955946445465, max_val=0.17587313055992126) 2025-09-09T15:08:41.8448402Z ) 2025-09-09T15:08:41.8448606Z (relu): ReLU(inplace=True) 2025-09-09T15:08:41.8448944Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:08:41.8449967Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:08:41.8451037Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.3842603862285614) 2025-09-09T15:08:41.8451506Z ) 2025-09-09T15:08:41.8451689Z ) 2025-09-09T15:08:41.8451788Z 2025-09-09T15:08:41.8451792Z 2025-09-09T15:08:41.8451796Z 2025-09-09T15:08:41.8451882Z def forward(self, x): 2025-09-09T15:08:41.8452243Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:08:41.8452678Z conv = self.conv(activation_post_process_0) 2025-09-09T15:08:41.8453121Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:08:41.8453825Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T15:08:41.8454389Z relu = self.relu(add); add = None 2025-09-09T15:08:41.8454811Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:08:41.8455235Z return activation_post_process_2 2025-09-09T15:08:41.8455502Z 2025-09-09T15:08:41.8455781Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:08:41.8456200Z diff: tensor([[[0., 0., 0.]]], grad_fn=) 2025-09-09T15:08:41.8456530Z converted model pt2e: GraphModule( 2025-09-09T15:08:41.8456799Z (conv): Module() 2025-09-09T15:08:41.8457002Z ) 2025-09-09T15:08:41.8457101Z 2025-09-09T15:08:41.8457114Z 2025-09-09T15:08:41.8457118Z 2025-09-09T15:08:41.8457209Z def forward(self, x): 2025-09-09T15:08:41.8457499Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:08:41.8457871Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T15:08:41.8458828Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0019342823652550578, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:08:41.8459689Z conv_bias = self.conv.bias 2025-09-09T15:08:41.8460326Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T15:08:41.8461448Z dequantize_per_tensor_default_5 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008883354254066944, 41, -128, 127, torch.int8) 2025-09-09T15:09:23.6564793Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:09:23.6566672Z conv1d = torch.ops.aten.conv1d.default(dequantize_per_tensor_default_4, dequantize_per_tensor_default, conv_bias); dequantize_per_tensor_default_4 = dequantize_per_tensor_default = conv_bias = None 2025-09-09T15:09:23.6568263Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d, 0.0021822303533554077, 46, -128, 127, torch.int8); conv1d = None 2025-09-09T15:09:23.6569908Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:09:23.6571580Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_2, dequantize_per_tensor_default_5); dequantize_per_tensor_default_2 = dequantize_per_tensor_default_5 = None 2025-09-09T15:09:23.6572965Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:09:23.6573913Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.0015069034416228533, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T15:09:23.6575392Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T15:09:23.6576415Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T15:09:23.6576829Z 2025-09-09T15:09:23.6577138Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:23.6577512Z onverted model fx: GraphModule( 2025-09-09T15:09:23.6577892Z (conv): QuantizedConv1d(Reference)(1, 1, kernel_size=(1,), stride=(1,)) 2025-09-09T15:09:23.6578288Z (relu): ReLU(inplace=True) 2025-09-09T15:09:23.6578530Z ) 2025-09-09T15:09:23.6578630Z 2025-09-09T15:09:23.6578634Z 2025-09-09T15:09:23.6578638Z 2025-09-09T15:09:23.6578726Z def forward(self, x): 2025-09-09T15:09:23.6579353Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.008883354254066944, 41, -128, 127, torch.int8); x = None 2025-09-09T15:09:23.6580599Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.008883354254066944, 41, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:09:23.6581481Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T15:09:23.6582214Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0021822303533554077, 46, -128, 127, torch.int8); conv = None 2025-09-09T15:09:23.6583490Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0021822303533554077, 46, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:09:23.6584701Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:09:23.6585341Z relu = self.relu(add); add = None 2025-09-09T15:09:23.6586033Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.0015069034416228533, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:09:23.6587325Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0015069034416228533, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:09:23.6588221Z return dequantize_per_tensor_default_2 2025-09-09T15:09:23.6588490Z 2025-09-09T15:09:23.6588773Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:23.6589142Z diff: tensor([[[0., 0., 0.]]]) 2025-09-09T15:09:23.6589599Z PASSED 2025-09-09T15:09:23.6590305Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_per_channel_weight_custom_dtype PASSED 2025-09-09T15:09:23.6591385Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_preserve_source_fn_stack PASSED 2025-09-09T15:09:23.6592354Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_update_shared_qspec model pt2e: GraphModule( 2025-09-09T15:09:23.6592965Z (conv): Module() 2025-09-09T15:09:23.6593170Z (bn): Module() 2025-09-09T15:09:23.6593467Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:23.6594411Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:23.6595614Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:09:23.6596115Z ) 2025-09-09T15:09:23.6596464Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:23.6597583Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0026]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:09:23.6598872Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3156, -0.1982, -0.3261]), max_val=tensor([0.2827, 0.2978, 0.2359])) 2025-09-09T15:09:23.6599517Z ) 2025-09-09T15:09:23.6599787Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:23.6600726Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:23.6601827Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T15:09:23.6602326Z ) 2025-09-09T15:09:23.6602606Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:23.6603534Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:23.6604626Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T15:09:23.6605159Z ) 2025-09-09T15:09:23.6605349Z ) 2025-09-09T15:09:23.6605447Z 2025-09-09T15:09:23.6605452Z 2025-09-09T15:09:23.6605461Z 2025-09-09T15:09:23.6605545Z def forward(self, x): 2025-09-09T15:09:23.6605833Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:09:23.6606173Z conv_weight = self.conv.weight 2025-09-09T15:09:23.6606442Z conv_bias = self.conv.bias 2025-09-09T15:09:23.6606704Z bn_weight = self.bn.weight 2025-09-09T15:09:23.6606967Z bn_bias = self.bn.bias 2025-09-09T15:09:23.6607228Z bn_running_mean = self.bn.running_mean 2025-09-09T15:09:23.6607538Z bn_running_var = self.bn.running_var 2025-09-09T15:09:23.6607869Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:09:23.6608311Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:09:23.6608901Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:09:23.6609435Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:09:23.6609831Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:09:23.6610242Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:09:23.6610693Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:09:23.6611194Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:09:23.6611760Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:09:23.6612377Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:09:23.6613370Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:09:23.6614256Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:09:23.6614786Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:09:23.6615363Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:09:23.6616016Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:09:23.6616886Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:09:23.6617895Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:09:23.6618642Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T15:09:23.6619359Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T15:09:23.6619941Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:09:23.6620328Z 2025-09-09T15:09:23.6620613Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:23.6620980Z model fx: GraphModule( 2025-09-09T15:09:23.6621309Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:23.6622259Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:23.6623366Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:09:23.6623877Z ) 2025-09-09T15:09:23.6624053Z (conv): ConvBn1d( 2025-09-09T15:09:44.3300099Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:09:44.3300649Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:09:44.3301255Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3302502Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0025, 0.0024, 0.0026]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:09:44.3304190Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.3156, -0.1982, -0.3261]), max_val=tensor([0.2827, 0.2978, 0.2359])) 2025-09-09T15:09:44.3305024Z ) 2025-09-09T15:09:44.3305270Z ) 2025-09-09T15:09:44.3305607Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3306797Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:44.3308179Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T15:09:44.3308822Z ) 2025-09-09T15:09:44.3309087Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:09:44.3309621Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3310797Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:44.3312177Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3043057918548584, max_val=1.399786114692688) 2025-09-09T15:09:44.3312814Z ) 2025-09-09T15:09:44.3313024Z ) 2025-09-09T15:09:44.3313142Z 2025-09-09T15:09:44.3313147Z 2025-09-09T15:09:44.3313152Z 2025-09-09T15:09:44.3313254Z def forward(self, x): 2025-09-09T15:09:44.3313680Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:09:44.3314325Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:09:44.3315001Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:09:44.3315998Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:09:44.3316839Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T15:09:44.3317402Z return activation_post_process_2 2025-09-09T15:09:44.3317716Z 2025-09-09T15:09:44.3318214Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:44.3318880Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:09:44.3319163Z [0., 0., 0.], 2025-09-09T15:09:44.3319447Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:09:44.3319810Z converted model pt2e: GraphModule( 2025-09-09T15:09:44.3320129Z (conv): Module() 2025-09-09T15:09:44.3320367Z (bn): Module() 2025-09-09T15:09:44.3320600Z ) 2025-09-09T15:09:44.3320716Z 2025-09-09T15:09:44.3320721Z 2025-09-09T15:09:44.3320726Z 2025-09-09T15:09:44.3320828Z def forward(self, x): 2025-09-09T15:09:44.3321170Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:09:44.3321576Z conv_bias = self.conv.bias 2025-09-09T15:09:44.3321945Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:09:44.3322809Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:09:44.3324329Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:09:44.3325609Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:09:44.3326174Z _scale_0 = self._scale_0 2025-09-09T15:09:44.3326485Z _zero_point_0 = self._zero_point_0 2025-09-09T15:09:44.3326855Z quantize_per_channel = self._frozen_param0 2025-09-09T15:09:44.3327931Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:09:44.3329618Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:09:44.3331138Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.010604282841086388, -5, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:09:44.3332770Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:09:44.3334241Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_1, -1.0, 1.0); dequantize_per_tensor_default_1 = None 2025-09-09T15:09:44.3335520Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010604282841086388, -5, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:09:44.3337253Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:09:44.3338292Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:09:44.3338694Z 2025-09-09T15:09:44.3338978Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:44.3339349Z onverted model fx: GraphModule( 2025-09-09T15:09:44.3339722Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:09:44.3340140Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:09:44.3340419Z ) 2025-09-09T15:09:44.3340515Z 2025-09-09T15:09:44.3340519Z 2025-09-09T15:09:44.3340523Z 2025-09-09T15:09:44.3340718Z def forward(self, x): 2025-09-09T15:09:44.3341315Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:09:44.3342609Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:09:44.3343612Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:09:44.3344453Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.010604282841086388, -5, -128, 127, torch.int8); conv = None 2025-09-09T15:09:44.3345712Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:09:44.3346773Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:09:44.3347685Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010604282841086388, -5, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:09:44.3348977Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010604282841086388, -5, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:09:44.3349842Z return dequantize_per_tensor_default_2 2025-09-09T15:09:44.3350116Z 2025-09-09T15:09:44.3350395Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:44.3350756Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:09:44.3350987Z [0., 0., 0.], 2025-09-09T15:09:44.3351190Z [0., 0., 0.]]]) 2025-09-09T15:09:44.3351418Z model pt2e: GraphModule( 2025-09-09T15:09:44.3351647Z (conv): Module() 2025-09-09T15:09:44.3351854Z (bn): Module() 2025-09-09T15:09:44.3352151Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3353088Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:44.3354179Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:09:44.3354686Z ) 2025-09-09T15:09:44.3354965Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3355896Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:09:44.3357051Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3261372148990631, max_val=0.297783762216568) 2025-09-09T15:09:44.3357558Z ) 2025-09-09T15:09:44.3357825Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3358755Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:44.3359831Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T15:09:44.3360335Z ) 2025-09-09T15:09:44.3360609Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:44.3361527Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:44.3362701Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T15:09:44.3363200Z ) 2025-09-09T15:09:44.3363372Z ) 2025-09-09T15:09:44.3363466Z 2025-09-09T15:09:44.3363471Z 2025-09-09T15:09:44.3363475Z 2025-09-09T15:09:44.3363565Z def forward(self, x): 2025-09-09T15:09:44.3364129Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:09:44.3364475Z conv_weight = self.conv.weight 2025-09-09T15:09:44.3364740Z conv_bias = self.conv.bias 2025-09-09T15:09:44.3364996Z bn_weight = self.bn.weight 2025-09-09T15:09:44.3365241Z bn_bias = self.bn.bias 2025-09-09T15:09:44.3365502Z bn_running_mean = self.bn.running_mean 2025-09-09T15:09:59.2065055Z bn_running_var = self.bn.running_var 2025-09-09T15:09:59.2065497Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:09:59.2066049Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:09:59.2066789Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:09:59.2067438Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:09:59.2067906Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:09:59.2068406Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:09:59.2068942Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1]) 2025-09-09T15:09:59.2069544Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:09:59.2070235Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:09:59.2070983Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:09:59.2072227Z conv1d_1 = torch.ops.aten.conv1d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:09:59.2073331Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1]); div = None 2025-09-09T15:09:59.2073976Z div_1 = torch.ops.aten.div.Tensor(conv1d_1, reshape_1); conv1d_1 = reshape_1 = None 2025-09-09T15:09:59.2074666Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1]); conv_bias = None 2025-09-09T15:09:59.2075341Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:09:59.2076485Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:09:59.2077657Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:09:59.2078586Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T15:09:59.2079467Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T15:09:59.2080183Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:09:59.2080656Z 2025-09-09T15:09:59.2080995Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:59.2081441Z model fx: GraphModule( 2025-09-09T15:09:59.2081837Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:59.2083018Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0104]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:59.2084414Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3264806270599365, max_val=1.318617343902588) 2025-09-09T15:09:59.2085055Z ) 2025-09-09T15:09:59.2085262Z (conv): ConvBn1d( 2025-09-09T15:09:59.2085528Z 3, 3, kernel_size=(3,), stride=(1,) 2025-09-09T15:09:59.2086346Z (bn): BatchNorm1d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:09:59.2086903Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:59.2088177Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0026]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:09:59.2089562Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.3261372148990631, max_val=0.297783762216568) 2025-09-09T15:09:59.2090208Z ) 2025-09-09T15:09:59.2090419Z ) 2025-09-09T15:09:59.2090739Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:59.2091921Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:59.2093308Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T15:09:59.2093943Z ) 2025-09-09T15:09:59.2094195Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:09:59.2094676Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:09:59.2095853Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0106]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:09:59.2097228Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.3146827220916748, max_val=1.399786114692688) 2025-09-09T15:09:59.2097859Z ) 2025-09-09T15:09:59.2098057Z ) 2025-09-09T15:09:59.2098177Z 2025-09-09T15:09:59.2098182Z 2025-09-09T15:09:59.2098187Z 2025-09-09T15:09:59.2098288Z def forward(self, x): 2025-09-09T15:09:59.2098711Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:09:59.2099369Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:09:59.2100033Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:09:59.2100744Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:09:59.2101502Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T15:09:59.2102064Z return activation_post_process_2 2025-09-09T15:09:59.2102370Z 2025-09-09T15:09:59.2102704Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:59.2103140Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:09:59.2103422Z [0., 0., 0.], 2025-09-09T15:09:59.2103694Z [0., 0., 0.]]], grad_fn=) 2025-09-09T15:09:59.2112500Z converted model pt2e: GraphModule( 2025-09-09T15:09:59.2112863Z (conv): Module() 2025-09-09T15:09:59.2113126Z (bn): Module() 2025-09-09T15:09:59.2113355Z ) 2025-09-09T15:09:59.2113478Z 2025-09-09T15:09:59.2113484Z 2025-09-09T15:09:59.2113489Z 2025-09-09T15:09:59.2113592Z def forward(self, x): 2025-09-09T15:09:59.2113931Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:09:59.2114353Z conv_bias = self.conv.bias 2025-09-09T15:09:59.2114729Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:09:59.2115602Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:09:59.2117288Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:09:59.2118673Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:09:59.2119366Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:09:59.2120158Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.002568009542301297, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:09:59.2121477Z conv1d_2 = torch.ops.aten.conv1d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:09:59.2122671Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1d_2, 0.010644976049661636, -4, -128, 127, torch.int8); conv1d_2 = None 2025-09-09T15:09:59.2123958Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:09:59.2125117Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_2, -1.0, 1.0); dequantize_per_tensor_default_2 = None 2025-09-09T15:09:59.2126140Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010644976049661636, -4, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:09:59.2127435Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T15:09:59.2128488Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T15:09:59.2128903Z 2025-09-09T15:09:59.2129180Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:09:59.2129557Z onverted model fx: GraphModule( 2025-09-09T15:09:59.2129927Z (conv): QuantizedConv1d(Reference)(3, 3, kernel_size=(3,), stride=(1,)) 2025-09-09T15:09:59.2130347Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:09:59.2130634Z ) 2025-09-09T15:09:59.2130742Z 2025-09-09T15:09:59.2130746Z 2025-09-09T15:09:59.2130750Z 2025-09-09T15:09:59.2130842Z def forward(self, x): 2025-09-09T15:09:59.2131452Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.010372933000326157, 0, -128, 127, torch.int8); x = None 2025-09-09T15:09:59.2132670Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.010372933000326157, 0, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:09:59.2133677Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:09:59.2134519Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.010644976049661636, -4, -128, 127, torch.int8); conv = None 2025-09-09T15:09:59.2135779Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:09:59.2136842Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:09:59.2137758Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.010644976049661636, -4, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:11:02.4271376Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.010644976049661636, -4, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:11:02.4272528Z return dequantize_per_tensor_default_2 2025-09-09T15:11:02.4272860Z 2025-09-09T15:11:02.4273206Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:02.4273660Z diff: tensor([[[0., 0., 0.], 2025-09-09T15:11:02.4273943Z [0., 0., 0.], 2025-09-09T15:11:02.4274537Z [0., 0., 0.]]]) 2025-09-09T15:11:02.4275044Z PASSED 2025-09-09T15:11:02.4275847Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_fold_bn_erases_bn_node PASSED 2025-09-09T15:11:02.4277454Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_bias_derived_qspec PASSED 2025-09-09T15:11:02.4278648Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion model pt2e: GraphModule( 2025-09-09T15:11:02.4279389Z (conv): Module() 2025-09-09T15:11:02.4279633Z (bn): Module() 2025-09-09T15:11:02.4279988Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:02.4281169Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:02.4282577Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:11:02.4283208Z ) 2025-09-09T15:11:02.4283536Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:02.4284758Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:11:02.4286347Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1894, -0.1850, -0.1857]), max_val=tensor([0.1520, 0.1622, 0.1895])) 2025-09-09T15:11:02.4287156Z ) 2025-09-09T15:11:02.4287480Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:02.4288625Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:02.4290001Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9920659065246582, max_val=2.112386465072632) 2025-09-09T15:11:02.4290626Z ) 2025-09-09T15:11:02.4290828Z ) 2025-09-09T15:11:02.4290943Z 2025-09-09T15:11:02.4290949Z 2025-09-09T15:11:02.4290963Z 2025-09-09T15:11:02.4291064Z def forward(self, x): 2025-09-09T15:11:02.4291404Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:11:02.4291819Z conv_weight = self.conv.weight 2025-09-09T15:11:02.4292142Z conv_bias = self.conv.bias 2025-09-09T15:11:02.4292452Z bn_weight = self.bn.weight 2025-09-09T15:11:02.4292746Z bn_bias = self.bn.bias 2025-09-09T15:11:02.4293054Z bn_running_mean = self.bn.running_mean 2025-09-09T15:11:02.4293410Z bn_running_var = self.bn.running_var 2025-09-09T15:11:02.4293831Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:11:02.4294375Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:11:02.4295088Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:11:02.4295729Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:11:02.4296200Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:11:02.4296701Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:11:02.4297241Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:11:02.4297858Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:11:02.4298569Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:11:02.4299325Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:11:02.4300559Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:11:02.4302824Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:11:02.4303489Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:11:02.4304297Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:11:02.4304985Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:11:02.4306076Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:11:02.4307249Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:11:02.4307989Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:11:02.4308479Z 2025-09-09T15:11:02.4308809Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:02.4309304Z model fx: GraphModule( 2025-09-09T15:11:02.4309684Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:02.4310875Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:02.4312269Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:11:02.4312898Z ) 2025-09-09T15:11:02.4313109Z (conv): ConvBn2d( 2025-09-09T15:11:02.4313371Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:11:02.4313871Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:11:02.4314440Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:02.4315652Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:11:02.4317389Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1894, -0.1850, -0.1857]), max_val=tensor([0.1520, 0.1622, 0.1895])) 2025-09-09T15:11:02.4318203Z ) 2025-09-09T15:11:02.4318405Z ) 2025-09-09T15:11:02.4318728Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:02.4319913Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:02.4321310Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9920659065246582, max_val=2.112386465072632) 2025-09-09T15:11:02.4321952Z ) 2025-09-09T15:11:02.4322149Z ) 2025-09-09T15:11:02.4322268Z 2025-09-09T15:11:02.4322273Z 2025-09-09T15:11:02.4322278Z 2025-09-09T15:11:02.4322377Z def forward(self, x): 2025-09-09T15:11:02.4322810Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:11:02.4323468Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:11:02.4324133Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:11:02.4324655Z return activation_post_process_1 2025-09-09T15:11:02.4324959Z 2025-09-09T15:11:02.4325312Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:02.4325789Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:11:02.4326089Z [0., 0., 0.], 2025-09-09T15:11:02.4326345Z [0., 0., 0.]], 2025-09-09T15:11:02.4326528Z 2025-09-09T15:11:02.4326713Z [[0., 0., 0.], 2025-09-09T15:11:02.4326917Z [0., 0., 0.], 2025-09-09T15:11:02.4327118Z [0., 0., 0.]], 2025-09-09T15:11:02.4327254Z 2025-09-09T15:11:02.4327335Z [[0., 0., 0.], 2025-09-09T15:11:02.4327531Z [0., 0., 0.], 2025-09-09T15:11:02.4327764Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:11:02.4328182Z converted model pt2e: GraphModule( 2025-09-09T15:11:02.4328445Z (conv): Module() 2025-09-09T15:11:02.4328637Z (bn): Module() 2025-09-09T15:11:02.4328826Z ) 2025-09-09T15:11:02.4328923Z 2025-09-09T15:11:02.4328927Z 2025-09-09T15:11:02.4328931Z 2025-09-09T15:11:02.4329017Z def forward(self, x): 2025-09-09T15:11:02.4329288Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:11:02.4329621Z conv_bias = self.conv.bias 2025-09-09T15:11:02.4330254Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:11:02.4331492Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:11:02.4332340Z _scale_0 = self._scale_0 2025-09-09T15:11:02.4332594Z _zero_point_0 = self._zero_point_0 2025-09-09T15:11:02.4332895Z quantize_per_channel = self._frozen_param0 2025-09-09T15:11:02.4333770Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:11:02.4335113Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:11:02.4336302Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01609589159488678, -4, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:11:02.4337574Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01609589159488678, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:11:02.4338576Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:11:02.4338982Z 2025-09-09T15:11:02.4339260Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:02.4339632Z onverted model fx: GraphModule( 2025-09-09T15:11:02.4340007Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:11:02.4340388Z ) 2025-09-09T15:11:02.4340484Z 2025-09-09T15:11:02.4340487Z 2025-09-09T15:11:02.4340491Z 2025-09-09T15:11:02.4340571Z def forward(self, x): 2025-09-09T15:11:23.3952772Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:11:23.3955546Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:11:23.3956987Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:11:23.3958059Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01609589159488678, -4, -128, 127, torch.int8); conv = None 2025-09-09T15:11:23.3959684Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01609589159488678, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:11:23.3960795Z return dequantize_per_tensor_default_1 2025-09-09T15:11:23.3961125Z 2025-09-09T15:11:23.3961811Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:23.3962257Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:11:23.3962567Z [0., 0., 0.], 2025-09-09T15:11:23.3962818Z [0., 0., 0.]], 2025-09-09T15:11:23.3962989Z 2025-09-09T15:11:23.3963087Z [[0., 0., 0.], 2025-09-09T15:11:23.3963359Z [0., 0., 0.], 2025-09-09T15:11:23.3963997Z [0., 0., 0.]], 2025-09-09T15:11:23.3964181Z 2025-09-09T15:11:23.3964280Z [[0., 0., 0.], 2025-09-09T15:11:23.3964528Z [0., 0., 0.], 2025-09-09T15:11:23.3964781Z [0., 0., 0.]]]]) 2025-09-09T15:11:23.3965061Z model pt2e: GraphModule( 2025-09-09T15:11:23.3965338Z (conv): Module() 2025-09-09T15:11:23.3965578Z (bn): Module() 2025-09-09T15:11:23.3965943Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:23.3967092Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:23.3968470Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:11:23.3969098Z ) 2025-09-09T15:11:23.3969426Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:23.3970599Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:11:23.3971982Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18937721848487854, max_val=0.18946029245853424) 2025-09-09T15:11:23.3972620Z ) 2025-09-09T15:11:23.3972951Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:23.3974098Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:23.3975464Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9922423362731934, max_val=2.1162424087524414) 2025-09-09T15:11:23.3976105Z ) 2025-09-09T15:11:23.3976306Z ) 2025-09-09T15:11:23.3976426Z 2025-09-09T15:11:23.3976432Z 2025-09-09T15:11:23.3976444Z 2025-09-09T15:11:23.3976545Z def forward(self, x): 2025-09-09T15:11:23.3976883Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:11:23.3977295Z conv_weight = self.conv.weight 2025-09-09T15:11:23.3977617Z conv_bias = self.conv.bias 2025-09-09T15:11:23.3977926Z bn_weight = self.bn.weight 2025-09-09T15:11:23.3978229Z bn_bias = self.bn.bias 2025-09-09T15:11:23.3978536Z bn_running_mean = self.bn.running_mean 2025-09-09T15:11:23.3978901Z bn_running_var = self.bn.running_var 2025-09-09T15:11:23.3979298Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:11:23.3979848Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:11:23.3980575Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:11:23.3981221Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:11:23.3981697Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:11:23.3982197Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:11:23.3982786Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:11:23.3983412Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:11:23.3984103Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:11:23.3984858Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:11:23.3986088Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:11:23.3987332Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:11:23.3988091Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:11:23.3988815Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:11:23.3989489Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:11:23.3990560Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:11:23.3991706Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:11:23.3992437Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:11:23.3992923Z 2025-09-09T15:11:23.3993255Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:23.3993697Z model fx: GraphModule( 2025-09-09T15:11:23.3994083Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:23.3995260Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:23.3996717Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:11:23.3997351Z ) 2025-09-09T15:11:23.3997568Z (conv): ConvBn2d( 2025-09-09T15:11:23.3997835Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:11:23.3998333Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:11:23.3998904Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:23.4000032Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:11:23.4001447Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18937721848487854, max_val=0.18946029245853424) 2025-09-09T15:11:23.4002095Z ) 2025-09-09T15:11:23.4002304Z ) 2025-09-09T15:11:23.4002629Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:23.4003812Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0161]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:23.4005200Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.9922423362731934, max_val=2.1162424087524414) 2025-09-09T15:11:23.4005844Z ) 2025-09-09T15:11:23.4006047Z ) 2025-09-09T15:11:23.4006179Z 2025-09-09T15:11:23.4006184Z 2025-09-09T15:11:23.4006190Z 2025-09-09T15:11:23.4006309Z def forward(self, x): 2025-09-09T15:11:23.4006777Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:11:23.4007476Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:11:23.4008019Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:11:23.4008448Z return activation_post_process_1 2025-09-09T15:11:23.4008708Z 2025-09-09T15:11:23.4008996Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:23.4009360Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:11:23.4009604Z [0., 0., 0.], 2025-09-09T15:11:23.4009816Z [0., 0., 0.]], 2025-09-09T15:11:23.4009961Z 2025-09-09T15:11:23.4010145Z [[0., 0., 0.], 2025-09-09T15:11:23.4010357Z [0., 0., 0.], 2025-09-09T15:11:23.4010562Z [0., 0., 0.]], 2025-09-09T15:11:23.4010700Z 2025-09-09T15:11:23.4010783Z [[0., 0., 0.], 2025-09-09T15:11:23.4010989Z [0., 0., 0.], 2025-09-09T15:11:23.4011230Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:11:23.4011620Z converted model pt2e: GraphModule( 2025-09-09T15:11:23.4011897Z (conv): Module() 2025-09-09T15:11:23.4012097Z (bn): Module() 2025-09-09T15:11:23.4012293Z ) 2025-09-09T15:11:23.4012394Z 2025-09-09T15:11:23.4012398Z 2025-09-09T15:11:23.4012402Z 2025-09-09T15:11:23.4012495Z def forward(self, x): 2025-09-09T15:11:23.4012772Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:11:23.4013106Z conv_bias = self.conv.bias 2025-09-09T15:11:23.4013741Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:11:23.4014979Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:11:23.4015857Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:11:23.4016650Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014918133383616805, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:11:23.4017903Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:11:23.4019089Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01611170545220375, -4, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:11:23.4020362Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01611170545220375, -4, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:11:23.4021373Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:11:23.4021781Z 2025-09-09T15:11:23.4022073Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:23.4022448Z onverted model fx: GraphModule( 2025-09-09T15:11:47.2533587Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:11:47.2534312Z ) 2025-09-09T15:11:47.2534497Z 2025-09-09T15:11:47.2534796Z 2025-09-09T15:11:47.2534803Z 2025-09-09T15:11:47.2534962Z def forward(self, x): 2025-09-09T15:11:47.2536100Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:11:47.2538348Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:11:47.2539503Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:11:47.2540368Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01611170545220375, -4, -128, 127, torch.int8); conv = None 2025-09-09T15:11:47.2541618Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01611170545220375, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:11:47.2542486Z return dequantize_per_tensor_default_1 2025-09-09T15:11:47.2542753Z 2025-09-09T15:11:47.2543038Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:47.2543442Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:11:47.2543674Z [0., 0., 0.], 2025-09-09T15:11:47.2545259Z [0., 0., 0.]], 2025-09-09T15:11:47.2545401Z 2025-09-09T15:11:47.2545481Z [[0., 0., 0.], 2025-09-09T15:11:47.2545691Z [0., 0., 0.], 2025-09-09T15:11:47.2545897Z [0., 0., 0.]], 2025-09-09T15:11:47.2546034Z 2025-09-09T15:11:47.2546108Z [[0., 0., 0.], 2025-09-09T15:11:47.2546482Z [0., 0., 0.], 2025-09-09T15:11:47.2546689Z [0., 0., 0.]]]]) 2025-09-09T15:11:47.2547134Z PASSED 2025-09-09T15:11:47.2547716Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion_cuda model pt2e: GraphModule( 2025-09-09T15:11:47.2548337Z (conv): Module() 2025-09-09T15:11:47.2548534Z (bn): Module() 2025-09-09T15:11:47.2548833Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:47.2549944Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:47.2551215Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:11:47.2551724Z ) 2025-09-09T15:11:47.2552004Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:47.2553160Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:11:47.2554691Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T15:11:47.2555411Z ) 2025-09-09T15:11:47.2555696Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:47.2556857Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0165], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:47.2558110Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1394810676574707, max_val=2.0564441680908203) 2025-09-09T15:11:47.2558622Z ) 2025-09-09T15:11:47.2558785Z ) 2025-09-09T15:11:47.2558889Z 2025-09-09T15:11:47.2558893Z 2025-09-09T15:11:47.2558897Z 2025-09-09T15:11:47.2558982Z def forward(self, x): 2025-09-09T15:11:47.2559266Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:11:47.2559598Z conv_weight = self.conv.weight 2025-09-09T15:11:47.2559872Z conv_bias = self.conv.bias 2025-09-09T15:11:47.2560121Z bn_weight = self.bn.weight 2025-09-09T15:11:47.2560377Z bn_bias = self.bn.bias 2025-09-09T15:11:47.2560633Z bn_running_mean = self.bn.running_mean 2025-09-09T15:11:47.2560933Z bn_running_var = self.bn.running_var 2025-09-09T15:11:47.2561257Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:11:47.2561701Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:11:47.2562288Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:11:47.2562806Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:11:47.2563201Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:11:47.2563607Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:11:47.2564219Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:11:47.2564722Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:11:47.2565415Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:11:47.2566027Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:11:47.2567087Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:11:47.2567973Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:11:47.2568552Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:11:47.2569129Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:11:47.2569685Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:11:47.2570540Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:11:47.2571466Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:11:47.2572052Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:11:47.2572442Z 2025-09-09T15:11:47.2572727Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:47.2573082Z model fx: GraphModule( 2025-09-09T15:11:47.2573402Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:47.2574504Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:47.2575753Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:11:47.2576260Z ) 2025-09-09T15:11:47.2576434Z (conv): ConvBn2d( 2025-09-09T15:11:47.2576659Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:11:47.2577060Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:11:47.2577529Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:47.2578641Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:11:47.2580178Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T15:11:47.2580909Z ) 2025-09-09T15:11:47.2581078Z ) 2025-09-09T15:11:47.2581354Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:11:47.2582455Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0165], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:11:47.2591338Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1394810676574707, max_val=2.0564441680908203) 2025-09-09T15:11:47.2591906Z ) 2025-09-09T15:11:47.2592083Z ) 2025-09-09T15:11:47.2592180Z 2025-09-09T15:11:47.2592184Z 2025-09-09T15:11:47.2592188Z 2025-09-09T15:11:47.2592274Z def forward(self, x): 2025-09-09T15:11:47.2592635Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:11:47.2593167Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:11:47.2593839Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:11:47.2594273Z return activation_post_process_1 2025-09-09T15:11:47.2594533Z 2025-09-09T15:11:47.2594822Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:11:47.2595267Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:11:47.2595514Z [0., 0., 0.], 2025-09-09T15:11:47.2595724Z [0., 0., 0.]], 2025-09-09T15:11:47.2595872Z 2025-09-09T15:11:47.2595949Z [[0., 0., 0.], 2025-09-09T15:11:47.2596157Z [0., 0., 0.], 2025-09-09T15:11:47.2596435Z [0., 0., 0.]], 2025-09-09T15:11:47.2596574Z 2025-09-09T15:11:47.2596664Z [[0., 0., 0.], 2025-09-09T15:11:47.2596871Z [0., 0., 0.], 2025-09-09T15:11:47.2597135Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T15:11:47.2597473Z converted model pt2e: GraphModule( 2025-09-09T15:11:47.2597737Z (conv): Module() 2025-09-09T15:11:47.2597941Z (bn): Module() 2025-09-09T15:11:47.2598153Z ) 2025-09-09T15:11:47.2598264Z 2025-09-09T15:11:47.2598270Z 2025-09-09T15:11:47.2598274Z 2025-09-09T15:11:47.2598372Z def forward(self, x): 2025-09-09T15:11:47.2598650Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:11:47.2598990Z conv_bias = self.conv.bias 2025-09-09T15:11:47.2599630Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:11:47.2600867Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:11:47.2601721Z _scale_0 = self._scale_0 2025-09-09T15:11:47.2601976Z _zero_point_0 = self._zero_point_0 2025-09-09T15:11:47.2602284Z quantize_per_channel = self._frozen_param0 2025-09-09T15:12:08.0642907Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:12:08.0645689Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:12:08.0646946Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.016454609110951424, 2, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:12:08.0648224Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016454609110951424, 2, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:12:08.0649219Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:12:08.0649644Z 2025-09-09T15:12:08.0649936Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:08.0650317Z onverted model fx: GraphModule( 2025-09-09T15:12:08.0650701Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:12:08.0651088Z ) 2025-09-09T15:12:08.0651187Z 2025-09-09T15:12:08.0651196Z 2025-09-09T15:12:08.0651200Z 2025-09-09T15:12:08.0651290Z def forward(self, x): 2025-09-09T15:12:08.0651907Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:12:08.0653150Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:12:08.0654160Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:12:08.0655352Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.016454609110951424, 2, -128, 127, torch.int8); conv = None 2025-09-09T15:12:08.0656754Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016454609110951424, 2, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:12:08.0657626Z return dequantize_per_tensor_default_1 2025-09-09T15:12:08.0657902Z 2025-09-09T15:12:08.0658178Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:08.0658551Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:12:08.0658789Z [0., 0., 0.], 2025-09-09T15:12:08.0658999Z [0., 0., 0.]], 2025-09-09T15:12:08.0659145Z 2025-09-09T15:12:08.0659226Z [[0., 0., 0.], 2025-09-09T15:12:08.0659431Z [0., 0., 0.], 2025-09-09T15:12:08.0659639Z [0., 0., 0.]], 2025-09-09T15:12:08.0659784Z 2025-09-09T15:12:08.0659859Z [[0., 0., 0.], 2025-09-09T15:12:08.0660066Z [0., 0., 0.], 2025-09-09T15:12:08.0660285Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T15:12:08.0660566Z model pt2e: GraphModule( 2025-09-09T15:12:08.0660800Z (conv): Module() 2025-09-09T15:12:08.0660999Z (bn): Module() 2025-09-09T15:12:08.0661309Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:08.0662421Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:08.0663863Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:12:08.0664372Z ) 2025-09-09T15:12:08.0664649Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:08.0665771Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:12:08.0667053Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T15:12:08.0667571Z ) 2025-09-09T15:12:08.0667848Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:08.0668944Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0164], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:08.0670204Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.137371778488159, max_val=2.0522286891937256) 2025-09-09T15:12:08.0670706Z ) 2025-09-09T15:12:08.0670875Z ) 2025-09-09T15:12:08.0670971Z 2025-09-09T15:12:08.0670975Z 2025-09-09T15:12:08.0670979Z 2025-09-09T15:12:08.0671070Z def forward(self, x): 2025-09-09T15:12:08.0671350Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:12:08.0671694Z conv_weight = self.conv.weight 2025-09-09T15:12:08.0671967Z conv_bias = self.conv.bias 2025-09-09T15:12:08.0672228Z bn_weight = self.bn.weight 2025-09-09T15:12:08.0672475Z bn_bias = self.bn.bias 2025-09-09T15:12:08.0672739Z bn_running_mean = self.bn.running_mean 2025-09-09T15:12:08.0673038Z bn_running_var = self.bn.running_var 2025-09-09T15:12:08.0673369Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:12:08.0673810Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:12:08.0674402Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:12:08.0675055Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:12:08.0675441Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:12:08.0675855Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:12:08.0676464Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:12:08.0676983Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:12:08.0677545Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:12:08.0678154Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:12:08.0679128Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:12:08.0680011Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:12:08.0680561Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:12:08.0681147Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:12:08.0681734Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:12:08.0682625Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:12:08.0683542Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:12:08.0684139Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:12:08.0684527Z 2025-09-09T15:12:08.0684806Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:08.0685171Z model fx: GraphModule( 2025-09-09T15:12:08.0685493Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:08.0686620Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:08.0687883Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:12:08.0688378Z ) 2025-09-09T15:12:08.0688561Z (conv): ConvBn2d( 2025-09-09T15:12:08.0688783Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:12:08.0689199Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:12:08.0689654Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:08.0690743Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:12:08.0692039Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T15:12:08.0692549Z ) 2025-09-09T15:12:08.0692725Z ) 2025-09-09T15:12:08.0692998Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:08.0694109Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0164], device='cuda:0'), zero_point=tensor([2], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:08.0695375Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.137371778488159, max_val=2.0522286891937256) 2025-09-09T15:12:08.0695966Z ) 2025-09-09T15:12:08.0696135Z ) 2025-09-09T15:12:08.0696231Z 2025-09-09T15:12:08.0696235Z 2025-09-09T15:12:08.0696239Z 2025-09-09T15:12:08.0696321Z def forward(self, x): 2025-09-09T15:12:08.0696674Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:12:08.0697325Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:12:08.0697868Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:12:08.0698295Z return activation_post_process_1 2025-09-09T15:12:08.0698550Z 2025-09-09T15:12:08.0698830Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:08.0699190Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:12:08.0699427Z [0., 0., 0.], 2025-09-09T15:12:08.0699632Z [0., 0., 0.]], 2025-09-09T15:12:08.0699779Z 2025-09-09T15:12:08.0699856Z [[0., 0., 0.], 2025-09-09T15:12:08.0700070Z [0., 0., 0.], 2025-09-09T15:12:08.0700273Z [0., 0., 0.]], 2025-09-09T15:12:08.0700409Z 2025-09-09T15:12:08.0700492Z [[0., 0., 0.], 2025-09-09T15:12:08.0700691Z [0., 0., 0.], 2025-09-09T15:12:08.0700958Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T15:12:08.0701289Z converted model pt2e: GraphModule( 2025-09-09T15:12:16.8149056Z (conv): Module() 2025-09-09T15:12:16.8149643Z (bn): Module() 2025-09-09T15:12:16.8149950Z ) 2025-09-09T15:12:16.8150081Z 2025-09-09T15:12:16.8150086Z 2025-09-09T15:12:16.8150123Z 2025-09-09T15:12:16.8150227Z def forward(self, x): 2025-09-09T15:12:16.8150569Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:12:16.8150982Z conv_bias = self.conv.bias 2025-09-09T15:12:16.8151791Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:12:16.8153381Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:12:16.8154503Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:12:16.8155505Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014933086931705475, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:12:16.8157177Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:12:16.8158695Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.016429806128144264, 2, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:12:16.8160329Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.016429806128144264, 2, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:12:16.8161617Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:12:16.8162138Z 2025-09-09T15:12:16.8162477Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:16.8162938Z onverted model fx: GraphModule( 2025-09-09T15:12:16.8163403Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:12:16.8164041Z ) 2025-09-09T15:12:16.8164159Z 2025-09-09T15:12:16.8164164Z 2025-09-09T15:12:16.8164169Z 2025-09-09T15:12:16.8164276Z def forward(self, x): 2025-09-09T15:12:16.8165047Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:12:16.8166631Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:12:16.8168157Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:12:16.8169359Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.016429806128144264, 2, -128, 127, torch.int8); conv = None 2025-09-09T15:12:16.8170939Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016429806128144264, 2, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:12:16.8172067Z return dequantize_per_tensor_default_1 2025-09-09T15:12:16.8172434Z 2025-09-09T15:12:16.8172752Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:16.8173119Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:12:16.8173350Z [0., 0., 0.], 2025-09-09T15:12:16.8173565Z [0., 0., 0.]], 2025-09-09T15:12:16.8173707Z 2025-09-09T15:12:16.8173788Z [[0., 0., 0.], 2025-09-09T15:12:16.8173986Z [0., 0., 0.], 2025-09-09T15:12:16.8174203Z [0., 0., 0.]], 2025-09-09T15:12:16.8174338Z 2025-09-09T15:12:16.8174410Z [[0., 0., 0.], 2025-09-09T15:12:16.8174614Z [0., 0., 0.], 2025-09-09T15:12:16.8174834Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T15:12:16.8175315Z PASSED 2025-09-09T15:12:16.8175924Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion_literal_args model pt2e: GraphModule( 2025-09-09T15:12:16.8176562Z (conv): Module() 2025-09-09T15:12:16.8176767Z (bn): Module() 2025-09-09T15:12:16.8177065Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:16.8178000Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:16.8179097Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T15:12:16.8179604Z ) 2025-09-09T15:12:16.8179877Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:16.8180859Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:12:16.8182136Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1897, -0.1774, -0.1913]), max_val=tensor([0.1806, 0.1870, 0.1478])) 2025-09-09T15:12:16.8182771Z ) 2025-09-09T15:12:16.8183049Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:16.8183996Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0279]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:16.8185073Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.481316328048706, max_val=3.622279405593872) 2025-09-09T15:12:16.8185579Z ) 2025-09-09T15:12:16.8185746Z ) 2025-09-09T15:12:16.8185849Z 2025-09-09T15:12:16.8185853Z 2025-09-09T15:12:16.8185857Z 2025-09-09T15:12:16.8185939Z def forward(self, x): 2025-09-09T15:12:16.8186219Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:12:16.8186548Z conv_weight = self.conv.weight 2025-09-09T15:12:16.8186816Z conv_bias = self.conv.bias 2025-09-09T15:12:16.8187061Z bn_weight = self.bn.weight 2025-09-09T15:12:16.8187308Z bn_bias = self.bn.bias 2025-09-09T15:12:16.8187557Z bn_running_mean = self.bn.running_mean 2025-09-09T15:12:16.8187851Z bn_running_var = self.bn.running_var 2025-09-09T15:12:16.8188273Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:12:16.8188704Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:12:16.8189286Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:12:16.8189879Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:12:16.8190272Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:12:16.8190672Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:12:16.8191113Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:12:16.8191607Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:12:16.8192159Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:12:16.8192763Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:12:16.8193735Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2, 2], [4, 4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:12:16.8194627Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:12:16.8195166Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:12:16.8195737Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:12:16.8196399Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:12:16.8197252Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:12:16.8198164Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:12:16.8198761Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:12:16.8199141Z 2025-09-09T15:12:16.8199417Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:16.8199768Z model fx: GraphModule( 2025-09-09T15:12:16.8200100Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:16.8201030Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:16.8202121Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T15:12:16.8202626Z ) 2025-09-09T15:12:16.8202799Z (conv): ConvBn2d( 2025-09-09T15:12:16.8203061Z 3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4) 2025-09-09T15:12:16.8203504Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:12:16.8203965Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:16.8204916Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:12:16.8206197Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1897, -0.1774, -0.1913]), max_val=tensor([0.1806, 0.1870, 0.1478])) 2025-09-09T15:12:16.8206840Z ) 2025-09-09T15:12:16.8207008Z ) 2025-09-09T15:12:16.8207284Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:16.8208257Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0279]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:16.8209432Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.481316328048706, max_val=3.622279405593872) 2025-09-09T15:12:16.8209932Z ) 2025-09-09T15:12:16.8210094Z ) 2025-09-09T15:12:16.8210191Z 2025-09-09T15:12:16.8210201Z 2025-09-09T15:12:16.8210205Z 2025-09-09T15:12:16.8210365Z def forward(self, x): 2025-09-09T15:12:16.8210707Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:12:16.8211230Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:12:16.8211768Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:12:37.5388989Z return activation_post_process_1 2025-09-09T15:12:37.5391447Z 2025-09-09T15:12:37.5391898Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:37.5392384Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5392739Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5393074Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5393369Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5393664Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5393957Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:12:37.5394171Z 2025-09-09T15:12:37.5394283Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5394573Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5394873Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5395172Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5395461Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5395756Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:12:37.5395958Z 2025-09-09T15:12:37.5396055Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5396462Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5396754Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5397051Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5397345Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5397702Z [0., 0., 0., 0., 0., 0.]]]], grad_fn=) 2025-09-09T15:12:37.5398095Z converted model pt2e: GraphModule( 2025-09-09T15:12:37.5398410Z (conv): Module() 2025-09-09T15:12:37.5398656Z (bn): Module() 2025-09-09T15:12:37.5398887Z ) 2025-09-09T15:12:37.5399012Z 2025-09-09T15:12:37.5399024Z 2025-09-09T15:12:37.5399029Z 2025-09-09T15:12:37.5399132Z def forward(self, x): 2025-09-09T15:12:37.5399470Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:12:37.5399885Z conv_bias = self.conv.bias 2025-09-09T15:12:37.5400696Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T15:12:37.5402251Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:12:37.5403311Z _scale_0 = self._scale_0 2025-09-09T15:12:37.5403614Z _zero_point_0 = self._zero_point_0 2025-09-09T15:12:37.5403979Z quantize_per_channel = self._frozen_param0 2025-09-09T15:12:37.5405060Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:12:37.5406736Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias, [2, 2], [4, 4]); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:12:37.5408239Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.027857238426804543, -3, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:12:37.5409827Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.027857238426804543, -3, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:12:37.5411340Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:12:37.5411842Z 2025-09-09T15:12:37.5412320Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:37.5412779Z onverted model fx: GraphModule( 2025-09-09T15:12:37.5413288Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4)) 2025-09-09T15:12:37.5413804Z ) 2025-09-09T15:12:37.5413922Z 2025-09-09T15:12:37.5413927Z 2025-09-09T15:12:37.5413932Z 2025-09-09T15:12:37.5414039Z def forward(self, x): 2025-09-09T15:12:37.5414796Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T15:12:37.5416325Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:12:37.5417567Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:12:37.5418615Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.027857238426804543, -3, -128, 127, torch.int8); conv = None 2025-09-09T15:12:37.5420288Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.027857238426804543, -3, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:12:37.5421204Z return dequantize_per_tensor_default_1 2025-09-09T15:12:37.5421482Z 2025-09-09T15:12:37.5421754Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:37.5422128Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5422405Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5422650Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5422889Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5423124Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5423368Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:12:37.5423536Z 2025-09-09T15:12:37.5423621Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5423871Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5424107Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5424353Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5424594Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5424831Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:12:37.5424995Z 2025-09-09T15:12:37.5425085Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5425320Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5425563Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5425798Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5426051Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:12:37.5426291Z [0., 0., 0., 0., 0., 0.]]]]) 2025-09-09T15:12:37.5426559Z model pt2e: GraphModule( 2025-09-09T15:12:37.5426796Z (conv): Module() 2025-09-09T15:12:37.5426999Z (bn): Module() 2025-09-09T15:12:37.5427307Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:37.5428245Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:37.5429390Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T15:12:37.5429895Z ) 2025-09-09T15:12:37.5430166Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:37.5431102Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:12:37.5432296Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19127479195594788, max_val=0.1870359182357788) 2025-09-09T15:12:37.5432807Z ) 2025-09-09T15:12:37.5433160Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:37.5434080Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0278]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:37.5435162Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.4796082973480225, max_val=3.620413064956665) 2025-09-09T15:12:37.5435653Z ) 2025-09-09T15:12:37.5435818Z ) 2025-09-09T15:12:37.5435918Z 2025-09-09T15:12:37.5435922Z 2025-09-09T15:12:37.5435926Z 2025-09-09T15:12:37.5436022Z def forward(self, x): 2025-09-09T15:12:37.5436363Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:12:37.5436697Z conv_weight = self.conv.weight 2025-09-09T15:12:37.5436964Z conv_bias = self.conv.bias 2025-09-09T15:12:37.5437221Z bn_weight = self.bn.weight 2025-09-09T15:12:37.5437463Z bn_bias = self.bn.bias 2025-09-09T15:12:37.5437728Z bn_running_mean = self.bn.running_mean 2025-09-09T15:12:37.5438023Z bn_running_var = self.bn.running_var 2025-09-09T15:12:37.5438352Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:12:37.5438789Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:12:37.5439368Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:12:37.5439890Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:12:37.5440278Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:12:37.5440690Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:12:37.5441130Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:12:37.5441635Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:12:37.5442192Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:12:37.5442799Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:12:37.5443779Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like, [2, 2], [4, 4]); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:12:37.5444668Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:12:37.5445208Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:12:37.5445794Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:12:37.5446348Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:12:37.5447215Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:12:37.5448126Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:12:37.5448718Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:12:37.5449119Z 2025-09-09T15:12:37.5449455Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:12:37.5449939Z model fx: GraphModule( 2025-09-09T15:12:37.5457816Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:12:37.5458781Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0165]), zero_point=tensor([-18], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:12:37.5460002Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.8076945543289185, max_val=2.388113498687744) 2025-09-09T15:12:37.5460508Z ) 2025-09-09T15:12:37.5460779Z (conv): ConvBn2d( 2025-09-09T15:12:37.5461050Z 3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4) 2025-09-09T15:13:05.5684573Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:13:05.5685169Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:05.5686332Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:13:05.5687768Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19127479195594788, max_val=0.1870359182357788) 2025-09-09T15:13:05.5688431Z ) 2025-09-09T15:13:05.5688643Z ) 2025-09-09T15:13:05.5688973Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:05.5690167Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0278]), zero_point=tensor([-3], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:05.5691568Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-3.4796082973480225, max_val=3.620413064956665) 2025-09-09T15:13:05.5692208Z ) 2025-09-09T15:13:05.5692428Z ) 2025-09-09T15:13:05.5692545Z 2025-09-09T15:13:05.5692570Z 2025-09-09T15:13:05.5692575Z 2025-09-09T15:13:05.5692689Z def forward(self, x): 2025-09-09T15:13:05.5693115Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:05.5693766Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:13:05.5694439Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:13:05.5694962Z return activation_post_process_1 2025-09-09T15:13:05.5695271Z 2025-09-09T15:13:05.5695610Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:05.5696068Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5696393Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5696697Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5696988Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5697289Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5697580Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:13:05.5697795Z 2025-09-09T15:13:05.5697892Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5698181Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5698473Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5698768Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5699071Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5699365Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:13:05.5699568Z 2025-09-09T15:13:05.5699663Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5699962Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5700248Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5700548Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5700837Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5701183Z [0., 0., 0., 0., 0., 0.]]]], grad_fn=) 2025-09-09T15:13:05.5701572Z converted model pt2e: GraphModule( 2025-09-09T15:13:05.5701889Z (conv): Module() 2025-09-09T15:13:05.5702135Z (bn): Module() 2025-09-09T15:13:05.5702362Z ) 2025-09-09T15:13:05.5702478Z 2025-09-09T15:13:05.5702484Z 2025-09-09T15:13:05.5702488Z 2025-09-09T15:13:05.5702596Z def forward(self, x): 2025-09-09T15:13:05.5702928Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:13:05.5703671Z conv_bias = self.conv.bias 2025-09-09T15:13:05.5704478Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T15:13:05.5706194Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:13:05.5707294Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:13:05.5708273Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0015061007579788566, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:13:05.5709862Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias, [2, 2], [4, 4]); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:13:05.5711375Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.02784322015941143, -3, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:13:05.5712978Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.02784322015941143, -3, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:13:05.5714229Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:13:05.5714736Z 2025-09-09T15:13:05.5715064Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:05.5715522Z onverted model fx: GraphModule( 2025-09-09T15:13:05.5716031Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(2, 2), padding=(4, 4)) 2025-09-09T15:13:05.5716708Z ) 2025-09-09T15:13:05.5716825Z 2025-09-09T15:13:05.5716830Z 2025-09-09T15:13:05.5716835Z 2025-09-09T15:13:05.5716937Z def forward(self, x): 2025-09-09T15:13:05.5717722Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.016454149037599564, -18, -128, 127, torch.int8); x = None 2025-09-09T15:13:05.5719318Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.016454149037599564, -18, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:13:05.5720606Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:13:05.5721744Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.02784322015941143, -3, -128, 127, torch.int8); conv = None 2025-09-09T15:13:05.5723077Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.02784322015941143, -3, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:13:05.5723938Z return dequantize_per_tensor_default_1 2025-09-09T15:13:05.5724216Z 2025-09-09T15:13:05.5724488Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:05.5724863Z diff: tensor([[[[0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5725125Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5725369Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5725612Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5725854Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5726101Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:13:05.5726266Z 2025-09-09T15:13:05.5726347Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5726591Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5726828Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5727068Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5727301Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5727542Z [0., 0., 0., 0., 0., 0.]], 2025-09-09T15:13:05.5727706Z 2025-09-09T15:13:05.5727881Z [[0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5728124Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5728371Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5728609Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5728877Z [0., 0., 0., 0., 0., 0.], 2025-09-09T15:13:05.5729226Z [0., 0., 0., 0., 0., 0.]]]]) 2025-09-09T15:13:05.5729702Z PASSED 2025-09-09T15:13:05.5730314Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T15:13:05.5730956Z (conv): Module() 2025-09-09T15:13:05.5731155Z (bn): Module() 2025-09-09T15:13:05.5731463Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:05.5732400Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:05.5733499Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:13:05.5734010Z ) 2025-09-09T15:13:05.5734280Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:05.5735268Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:13:05.5736546Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T15:13:05.5737182Z ) 2025-09-09T15:13:05.5737462Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:05.5738385Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:05.5739476Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1459269523620605, max_val=2.376943349838257) 2025-09-09T15:13:05.5739981Z ) 2025-09-09T15:13:05.5740152Z ) 2025-09-09T15:13:05.5740259Z 2025-09-09T15:13:05.5740270Z 2025-09-09T15:13:05.5740274Z 2025-09-09T15:13:05.5740357Z def forward(self, x): 2025-09-09T15:13:05.5740641Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:13:05.5740978Z conv_weight = self.conv.weight 2025-09-09T15:13:05.5741248Z bn_weight = self.bn.weight 2025-09-09T15:13:05.5741495Z bn_bias = self.bn.bias 2025-09-09T15:13:05.5741760Z bn_running_mean = self.bn.running_mean 2025-09-09T15:13:05.5742052Z bn_running_var = self.bn.running_var 2025-09-09T15:13:05.5742380Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:13:05.5742808Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:05.5743400Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:13:05.5743921Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:13:05.5744311Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:13:05.5744722Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:13:05.5745157Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:13:05.5745664Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:13:05.5746217Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:13:05.5747048Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:13:05.5747868Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:13:05.5748609Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:13:23.5805165Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:13:23.5806383Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:13:23.5807128Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:13:23.5807607Z 2025-09-09T15:13:23.5807950Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:23.5808389Z model fx: GraphModule( 2025-09-09T15:13:23.5808791Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:23.5809960Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:23.5811424Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:13:23.5812074Z ) 2025-09-09T15:13:23.5812301Z (conv): ConvBn2d( 2025-09-09T15:13:23.5812597Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:13:23.5813128Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:13:23.5813693Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:23.5814900Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:13:23.5816546Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T15:13:23.5817367Z ) 2025-09-09T15:13:23.5817578Z ) 2025-09-09T15:13:23.5817905Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:23.5819070Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:23.5820438Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1459269523620605, max_val=2.376943349838257) 2025-09-09T15:13:23.5821062Z ) 2025-09-09T15:13:23.5821268Z ) 2025-09-09T15:13:23.5821385Z 2025-09-09T15:13:23.5821390Z 2025-09-09T15:13:23.5821395Z 2025-09-09T15:13:23.5821497Z def forward(self, x): 2025-09-09T15:13:23.5821927Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:23.5822584Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:13:23.5823264Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:13:23.5823794Z return activation_post_process_1 2025-09-09T15:13:23.5824106Z 2025-09-09T15:13:23.5824448Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:23.5824897Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:13:23.5825191Z [0., 0., 0.], 2025-09-09T15:13:23.5825441Z [0., 0., 0.]], 2025-09-09T15:13:23.5825618Z 2025-09-09T15:13:23.5825711Z [[0., 0., 0.], 2025-09-09T15:13:23.5825956Z [0., 0., 0.], 2025-09-09T15:13:23.5826214Z [0., 0., 0.]], 2025-09-09T15:13:23.5826384Z 2025-09-09T15:13:23.5826481Z [[0., 0., 0.], 2025-09-09T15:13:23.5826724Z [0., 0., 0.], 2025-09-09T15:13:23.5826991Z [0., 0., 0.]]], 2025-09-09T15:13:23.5827161Z 2025-09-09T15:13:23.5827345Z 2025-09-09T15:13:23.5827448Z [[[0., 0., 0.], 2025-09-09T15:13:23.5827696Z [0., 0., 0.], 2025-09-09T15:13:23.5827948Z [0., 0., 0.]], 2025-09-09T15:13:23.5828113Z 2025-09-09T15:13:23.5828201Z [[0., 0., 0.], 2025-09-09T15:13:23.5828455Z [0., 0., 0.], 2025-09-09T15:13:23.5828697Z [0., 0., 0.]], 2025-09-09T15:13:23.5828958Z 2025-09-09T15:13:23.5829050Z [[0., 0., 0.], 2025-09-09T15:13:23.5829308Z [0., 0., 0.], 2025-09-09T15:13:23.5829554Z [0., 0., 0.]]], 2025-09-09T15:13:23.5829721Z 2025-09-09T15:13:23.5829726Z 2025-09-09T15:13:23.5829822Z [[[0., 0., 0.], 2025-09-09T15:13:23.5830065Z [0., 0., 0.], 2025-09-09T15:13:23.5830315Z [0., 0., 0.]], 2025-09-09T15:13:23.5830480Z 2025-09-09T15:13:23.5830569Z [[0., 0., 0.], 2025-09-09T15:13:23.5830815Z [0., 0., 0.], 2025-09-09T15:13:23.5831057Z [0., 0., 0.]], 2025-09-09T15:13:23.5831229Z 2025-09-09T15:13:23.5831317Z [[0., 0., 0.], 2025-09-09T15:13:23.5831577Z [0., 0., 0.], 2025-09-09T15:13:23.5831855Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:13:23.5832232Z converted model pt2e: GraphModule( 2025-09-09T15:13:23.5832543Z (conv): Module() 2025-09-09T15:13:23.5832786Z (bn): Module() 2025-09-09T15:13:23.5833013Z ) 2025-09-09T15:13:23.5833132Z 2025-09-09T15:13:23.5833144Z 2025-09-09T15:13:23.5833149Z 2025-09-09T15:13:23.5833248Z def forward(self, x): 2025-09-09T15:13:23.5833578Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:13:23.5834465Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:13:23.5836030Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:13:23.5837218Z _scale_0 = self._scale_0 2025-09-09T15:13:23.5837529Z _zero_point_0 = self._zero_point_0 2025-09-09T15:13:23.5837889Z quantize_per_channel = self._frozen_param0 2025-09-09T15:13:23.5838988Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:13:23.5840081Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:13:23.5841164Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T15:13:23.5842768Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01773674599826336, -7, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:13:23.5844406Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01773674599826336, -7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:13:23.5845682Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:13:23.5846197Z 2025-09-09T15:13:23.5846534Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:23.5847019Z onverted model fx: GraphModule( 2025-09-09T15:13:23.5847507Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:13:23.5848005Z ) 2025-09-09T15:13:23.5848118Z 2025-09-09T15:13:23.5848123Z 2025-09-09T15:13:23.5848126Z 2025-09-09T15:13:23.5848216Z def forward(self, x): 2025-09-09T15:13:23.5848822Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:13:23.5850040Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:13:23.5851166Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:13:23.5852076Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01773674599826336, -7, -128, 127, torch.int8); conv = None 2025-09-09T15:13:23.5853321Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01773674599826336, -7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:13:23.5854185Z return dequantize_per_tensor_default_1 2025-09-09T15:13:23.5854457Z 2025-09-09T15:13:23.5854737Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:23.5855096Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:13:23.5855333Z [0., 0., 0.], 2025-09-09T15:13:23.5855544Z [0., 0., 0.]], 2025-09-09T15:13:23.5855683Z 2025-09-09T15:13:23.5855764Z [[0., 0., 0.], 2025-09-09T15:13:23.5855966Z [0., 0., 0.], 2025-09-09T15:13:23.5856171Z [0., 0., 0.]], 2025-09-09T15:13:23.5856307Z 2025-09-09T15:13:23.5856381Z [[0., 0., 0.], 2025-09-09T15:13:23.5856593Z [0., 0., 0.], 2025-09-09T15:13:23.5856804Z [0., 0., 0.]]], 2025-09-09T15:13:23.5856953Z 2025-09-09T15:13:23.5856957Z 2025-09-09T15:13:23.5857031Z [[[0., 0., 0.], 2025-09-09T15:13:23.5857238Z [0., 0., 0.], 2025-09-09T15:13:23.5857437Z [0., 0., 0.]], 2025-09-09T15:13:23.5857574Z 2025-09-09T15:13:23.5857654Z [[0., 0., 0.], 2025-09-09T15:13:23.5857851Z [0., 0., 0.], 2025-09-09T15:13:23.5858061Z [0., 0., 0.]], 2025-09-09T15:13:23.5858199Z 2025-09-09T15:13:23.5858275Z [[0., 0., 0.], 2025-09-09T15:13:23.5858481Z [0., 0., 0.], 2025-09-09T15:13:23.5858680Z [0., 0., 0.]]], 2025-09-09T15:13:23.5858832Z 2025-09-09T15:13:23.5858836Z 2025-09-09T15:13:23.5858911Z [[[0., 0., 0.], 2025-09-09T15:13:23.5859116Z [0., 0., 0.], 2025-09-09T15:13:23.5859314Z [0., 0., 0.]], 2025-09-09T15:13:23.5859455Z 2025-09-09T15:13:23.5859527Z [[0., 0., 0.], 2025-09-09T15:13:23.5859724Z [0., 0., 0.], 2025-09-09T15:13:23.5859937Z [0., 0., 0.]], 2025-09-09T15:13:23.5860072Z 2025-09-09T15:13:23.5860144Z [[0., 0., 0.], 2025-09-09T15:13:23.5860344Z [0., 0., 0.], 2025-09-09T15:13:23.5860551Z [0., 0., 0.]]]]) 2025-09-09T15:13:23.5860777Z model pt2e: GraphModule( 2025-09-09T15:13:23.5861004Z (conv): Module() 2025-09-09T15:13:23.5861202Z (bn): Module() 2025-09-09T15:13:23.5861499Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:23.5862430Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:23.5863524Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:13:23.5864260Z ) 2025-09-09T15:13:23.5864530Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:23.5865467Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:13:23.5866559Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T15:13:23.5867068Z ) 2025-09-09T15:13:23.5867337Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:23.5868259Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:23.5869469Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1475415229797363, max_val=2.368046283721924) 2025-09-09T15:13:23.5869966Z ) 2025-09-09T15:13:23.5870132Z ) 2025-09-09T15:13:23.5870228Z 2025-09-09T15:13:23.5870347Z 2025-09-09T15:13:23.5870352Z 2025-09-09T15:13:23.5870460Z def forward(self, x): 2025-09-09T15:13:23.5870765Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:13:23.5871098Z conv_weight = self.conv.weight 2025-09-09T15:13:23.5871364Z bn_weight = self.bn.weight 2025-09-09T15:13:23.5871616Z bn_bias = self.bn.bias 2025-09-09T15:13:23.5871866Z bn_running_mean = self.bn.running_mean 2025-09-09T15:13:23.5872162Z bn_running_var = self.bn.running_var 2025-09-09T15:13:23.5872489Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:13:44.5387920Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:44.5390567Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:13:44.5391264Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:13:44.5391743Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:13:44.5392253Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:13:44.5392793Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:13:44.5393411Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:13:44.5394098Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:13:44.5395151Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:13:44.5396300Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:13:44.5396967Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:13:44.5398080Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:13:44.5399278Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:13:44.5400026Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:13:44.5400501Z 2025-09-09T15:13:44.5400844Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:44.5401286Z model fx: GraphModule( 2025-09-09T15:13:44.5401680Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5402860Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5404330Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:13:44.5404974Z ) 2025-09-09T15:13:44.5405190Z (conv): ConvBn2d( 2025-09-09T15:13:44.5405489Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:13:44.5406009Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:13:44.5406568Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5407696Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:13:44.5409070Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T15:13:44.5409972Z ) 2025-09-09T15:13:44.5410174Z ) 2025-09-09T15:13:44.5410506Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5411812Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0177]), zero_point=tensor([-7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5413136Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1475415229797363, max_val=2.368046283721924) 2025-09-09T15:13:44.5413650Z ) 2025-09-09T15:13:44.5413816Z ) 2025-09-09T15:13:44.5413916Z 2025-09-09T15:13:44.5413920Z 2025-09-09T15:13:44.5413924Z 2025-09-09T15:13:44.5414010Z def forward(self, x): 2025-09-09T15:13:44.5414369Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:44.5414900Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:13:44.5415457Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:13:44.5415880Z return activation_post_process_1 2025-09-09T15:13:44.5416144Z 2025-09-09T15:13:44.5416418Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:44.5416799Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:13:44.5417032Z [0., 0., 0.], 2025-09-09T15:13:44.5417244Z [0., 0., 0.]], 2025-09-09T15:13:44.5417386Z 2025-09-09T15:13:44.5417467Z [[0., 0., 0.], 2025-09-09T15:13:44.5417671Z [0., 0., 0.], 2025-09-09T15:13:44.5417882Z [0., 0., 0.]], 2025-09-09T15:13:44.5418021Z 2025-09-09T15:13:44.5418095Z [[0., 0., 0.], 2025-09-09T15:13:44.5418299Z [0., 0., 0.], 2025-09-09T15:13:44.5418502Z [0., 0., 0.]]], 2025-09-09T15:13:44.5418649Z 2025-09-09T15:13:44.5418676Z 2025-09-09T15:13:44.5418753Z [[[0., 0., 0.], 2025-09-09T15:13:44.5418963Z [0., 0., 0.], 2025-09-09T15:13:44.5419163Z [0., 0., 0.]], 2025-09-09T15:13:44.5419308Z 2025-09-09T15:13:44.5419383Z [[0., 0., 0.], 2025-09-09T15:13:44.5419582Z [0., 0., 0.], 2025-09-09T15:13:44.5419842Z [0., 0., 0.]], 2025-09-09T15:13:44.5419986Z 2025-09-09T15:13:44.5420067Z [[0., 0., 0.], 2025-09-09T15:13:44.5420275Z [0., 0., 0.], 2025-09-09T15:13:44.5420481Z [0., 0., 0.]]], 2025-09-09T15:13:44.5420623Z 2025-09-09T15:13:44.5420627Z 2025-09-09T15:13:44.5420700Z [[[0., 0., 0.], 2025-09-09T15:13:44.5420908Z [0., 0., 0.], 2025-09-09T15:13:44.5421109Z [0., 0., 0.]], 2025-09-09T15:13:44.5421253Z 2025-09-09T15:13:44.5421327Z [[0., 0., 0.], 2025-09-09T15:13:44.5421524Z [0., 0., 0.], 2025-09-09T15:13:44.5421731Z [0., 0., 0.]], 2025-09-09T15:13:44.5421865Z 2025-09-09T15:13:44.5421944Z [[0., 0., 0.], 2025-09-09T15:13:44.5422144Z [0., 0., 0.], 2025-09-09T15:13:44.5422392Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:13:44.5422701Z converted model pt2e: GraphModule( 2025-09-09T15:13:44.5422967Z (conv): Module() 2025-09-09T15:13:44.5423168Z (bn): Module() 2025-09-09T15:13:44.5423360Z ) 2025-09-09T15:13:44.5423459Z 2025-09-09T15:13:44.5423462Z 2025-09-09T15:13:44.5423466Z 2025-09-09T15:13:44.5423554Z def forward(self, x): 2025-09-09T15:13:44.5423836Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:13:44.5424564Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:13:44.5425798Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:13:44.5426676Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:13:44.5427564Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014933086931705475, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:13:44.5428367Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:13:44.5429285Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T15:13:44.5430533Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.01770818792283535, -7, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:13:44.5431821Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.01770818792283535, -7, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:13:44.5432829Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:13:44.5433262Z 2025-09-09T15:13:44.5433545Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:44.5434013Z onverted model fx: GraphModule( 2025-09-09T15:13:44.5442362Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:13:44.5442752Z ) 2025-09-09T15:13:44.5442866Z 2025-09-09T15:13:44.5442871Z 2025-09-09T15:13:44.5442875Z 2025-09-09T15:13:44.5442963Z def forward(self, x): 2025-09-09T15:13:44.5443592Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:13:44.5444873Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:13:44.5445884Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:13:44.5446745Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.01770818792283535, -7, -128, 127, torch.int8); conv = None 2025-09-09T15:13:44.5448004Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01770818792283535, -7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:13:44.5448884Z return dequantize_per_tensor_default_1 2025-09-09T15:13:44.5449154Z 2025-09-09T15:13:44.5449441Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:44.5449810Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:13:44.5450055Z [0., 0., 0.], 2025-09-09T15:13:44.5450275Z [0., 0., 0.]], 2025-09-09T15:13:44.5450416Z 2025-09-09T15:13:44.5450495Z [[0., 0., 0.], 2025-09-09T15:13:44.5450710Z [0., 0., 0.], 2025-09-09T15:13:44.5450913Z [0., 0., 0.]], 2025-09-09T15:13:44.5451062Z 2025-09-09T15:13:44.5451139Z [[0., 0., 0.], 2025-09-09T15:13:44.5451344Z [0., 0., 0.], 2025-09-09T15:13:44.5451549Z [0., 0., 0.]]], 2025-09-09T15:13:44.5451690Z 2025-09-09T15:13:44.5451694Z 2025-09-09T15:13:44.5451774Z [[[0., 0., 0.], 2025-09-09T15:13:44.5451976Z [0., 0., 0.], 2025-09-09T15:13:44.5452196Z [0., 0., 0.]], 2025-09-09T15:13:44.5452334Z 2025-09-09T15:13:44.5452412Z [[0., 0., 0.], 2025-09-09T15:13:44.5452620Z [0., 0., 0.], 2025-09-09T15:13:44.5452821Z [0., 0., 0.]], 2025-09-09T15:13:44.5452968Z 2025-09-09T15:13:44.5453041Z [[0., 0., 0.], 2025-09-09T15:13:44.5453253Z [0., 0., 0.], 2025-09-09T15:13:44.5453455Z [0., 0., 0.]]], 2025-09-09T15:13:44.5453595Z 2025-09-09T15:13:44.5453599Z 2025-09-09T15:13:44.5453683Z [[[0., 0., 0.], 2025-09-09T15:13:44.5453887Z [0., 0., 0.], 2025-09-09T15:13:44.5454095Z [0., 0., 0.]], 2025-09-09T15:13:44.5454352Z 2025-09-09T15:13:44.5454426Z [[0., 0., 0.], 2025-09-09T15:13:44.5454636Z [0., 0., 0.], 2025-09-09T15:13:44.5454837Z [0., 0., 0.]], 2025-09-09T15:13:44.5454979Z 2025-09-09T15:13:44.5455053Z [[0., 0., 0.], 2025-09-09T15:13:44.5455260Z [0., 0., 0.], 2025-09-09T15:13:44.5455547Z [0., 0., 0.]]]]) 2025-09-09T15:13:44.5455789Z model pt2e: GraphModule( 2025-09-09T15:13:44.5456016Z (conv1): Module() 2025-09-09T15:13:44.5456223Z (bn1): Module() 2025-09-09T15:13:44.5456418Z (conv2): Module() 2025-09-09T15:13:44.5456619Z (bn2): Module() 2025-09-09T15:13:44.5456915Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5457868Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5458976Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:13:44.5459491Z ) 2025-09-09T15:13:44.5459769Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5460758Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:13:44.5462072Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1921, -0.1899, -0.1895]), max_val=tensor([0.1769, 0.1726, 0.1697])) 2025-09-09T15:13:44.5462715Z ) 2025-09-09T15:13:44.5462987Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5464272Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:13:44.5465556Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T15:13:44.5466191Z ) 2025-09-09T15:13:44.5466473Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5467409Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5468499Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7323514223098755, max_val=2.7138354778289795) 2025-09-09T15:13:44.5469010Z ) 2025-09-09T15:13:44.5469281Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5470208Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5471301Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4053945541381836, max_val=1.4082176685333252) 2025-09-09T15:13:44.5471806Z ) 2025-09-09T15:13:44.5471980Z ) 2025-09-09T15:13:44.5472076Z 2025-09-09T15:13:44.5472080Z 2025-09-09T15:13:44.5472083Z 2025-09-09T15:13:44.5472166Z def forward(self, x): 2025-09-09T15:13:44.5472455Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:13:44.5472788Z conv1_weight = self.conv1.weight 2025-09-09T15:13:44.5473075Z bn1_weight = self.bn1.weight 2025-09-09T15:13:44.5473335Z bn1_bias = self.bn1.bias 2025-09-09T15:13:44.5473589Z conv2_weight = self.conv2.weight 2025-09-09T15:13:44.5473873Z conv2_bias = self.conv2.bias 2025-09-09T15:13:44.5474127Z bn2_weight = self.bn2.weight 2025-09-09T15:13:44.5474411Z bn2_bias = self.bn2.bias 2025-09-09T15:13:44.5474848Z bn1_running_mean = self.bn1.running_mean 2025-09-09T15:13:44.5475159Z bn1_running_var = self.bn1.running_var 2025-09-09T15:13:44.5475499Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T15:13:44.5475852Z bn2_running_mean = self.bn2.running_mean 2025-09-09T15:13:44.5476329Z bn2_running_var = self.bn2.running_var 2025-09-09T15:13:44.5476663Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T15:13:44.5477109Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:44.5477699Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T15:13:44.5478371Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T15:13:44.5478909Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T15:13:44.5479307Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:13:44.5479734Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T15:13:44.5480181Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:13:44.5480698Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T15:13:44.5481268Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T15:13:44.5481884Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:13:44.5482433Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T15:13:44.5482846Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T15:13:44.5483283Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T15:13:44.5483743Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1, 1]) 2025-09-09T15:13:44.5484278Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T15:13:44.5484869Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T15:13:44.5485721Z conv2d_3 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:13:44.5486559Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1, 1]); div_2 = None 2025-09-09T15:13:44.5487200Z div_3 = torch.ops.aten.div.Tensor(conv2d_3, reshape_4); conv2d_3 = reshape_4 = None 2025-09-09T15:13:44.5488319Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T15:13:44.5489478Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T15:13:44.5490644Z conv2d_2 = torch.ops.aten.conv2d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T15:13:44.5491724Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:13:44.5492362Z div_1 = torch.ops.aten.div.Tensor(conv2d_2, reshape_1); conv2d_2 = reshape_1 = None 2025-09-09T15:13:44.5493056Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1, 1]); conv2_bias = None 2025-09-09T15:13:44.5493784Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:13:44.5495135Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T15:13:44.5496485Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T15:13:44.5497278Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T15:13:44.5497724Z 2025-09-09T15:13:44.5498026Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:44.5498439Z model fx: GraphModule( 2025-09-09T15:13:44.5498789Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5500012Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5501117Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:13:44.5501622Z ) 2025-09-09T15:13:44.5501804Z (conv1): ConvBn2d( 2025-09-09T15:13:44.5502056Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:13:44.5502491Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:13:44.5502959Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5504042Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:13:44.5505637Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T15:13:44.5506279Z ) 2025-09-09T15:13:44.5506456Z ) 2025-09-09T15:13:44.5506735Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5507669Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5508777Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7323514223098755, max_val=2.7138354778289795) 2025-09-09T15:13:44.5509281Z ) 2025-09-09T15:13:44.5509462Z (conv2): ConvBn2d( 2025-09-09T15:13:44.5509693Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:13:44.5510109Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:13:44.5510577Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5511532Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:13:44.5512825Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1921, -0.1899, -0.1895]), max_val=tensor([0.1769, 0.1726, 0.1697])) 2025-09-09T15:13:44.5513472Z ) 2025-09-09T15:13:44.5513642Z ) 2025-09-09T15:13:44.5513920Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:13:44.5514856Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:13:44.5515956Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4053945541381836, max_val=1.4082176685333252) 2025-09-09T15:13:44.5516533Z ) 2025-09-09T15:13:44.5516709Z ) 2025-09-09T15:13:44.5516805Z 2025-09-09T15:13:44.5516810Z 2025-09-09T15:13:44.5516814Z 2025-09-09T15:13:44.5516904Z def forward(self, x): 2025-09-09T15:13:44.5517251Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:13:44.5517792Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:13:44.5518341Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T15:13:44.5518990Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:13:44.5519536Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T15:13:44.5519971Z return activation_post_process_2 2025-09-09T15:13:44.5520233Z 2025-09-09T15:13:44.5520615Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:13:44.5520977Z diff: tensor([[[[0.]], 2025-09-09T15:13:44.5521117Z 2025-09-09T15:13:44.5521191Z [[0.]], 2025-09-09T15:13:44.5521316Z 2025-09-09T15:14:02.6053352Z [[0.]]], 2025-09-09T15:14:02.6053564Z 2025-09-09T15:14:02.6053570Z 2025-09-09T15:14:02.6053663Z [[[0.]], 2025-09-09T15:14:02.6053809Z 2025-09-09T15:14:02.6053904Z [[0.]], 2025-09-09T15:14:02.6054047Z 2025-09-09T15:14:02.6054138Z [[0.]]], 2025-09-09T15:14:02.6054287Z 2025-09-09T15:14:02.6054292Z 2025-09-09T15:14:02.6054416Z [[[0.]], 2025-09-09T15:14:02.6054557Z 2025-09-09T15:14:02.6054645Z [[0.]], 2025-09-09T15:14:02.6054796Z 2025-09-09T15:14:02.6054918Z [[0.]]]], grad_fn=) 2025-09-09T15:14:02.6055290Z converted model pt2e: GraphModule( 2025-09-09T15:14:02.6055610Z (conv1): Module() 2025-09-09T15:14:02.6055862Z (bn1): Module() 2025-09-09T15:14:02.6056113Z (conv2): Module() 2025-09-09T15:14:02.6056360Z (bn2): Module() 2025-09-09T15:14:02.6056589Z ) 2025-09-09T15:14:02.6056711Z 2025-09-09T15:14:02.6056728Z 2025-09-09T15:14:02.6056733Z 2025-09-09T15:14:02.6056837Z def forward(self, x): 2025-09-09T15:14:02.6057175Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:14:02.6057593Z conv2_bias = self.conv2.bias 2025-09-09T15:14:02.6058415Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:14:02.6059983Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:14:02.6061073Z _scale_0 = self._scale_0 2025-09-09T15:14:02.6061379Z _zero_point_0 = self._zero_point_0 2025-09-09T15:14:02.6061719Z _scale_1 = self._scale_1 2025-09-09T15:14:02.6062026Z _zero_point_1 = self._zero_point_1 2025-09-09T15:14:02.6062396Z quantize_per_channel_1 = self._frozen_param0 2025-09-09T15:14:02.6063509Z dequantize_per_channel_1 = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_1, _scale_1, _zero_point_1, 0, -127, 127, torch.int8); quantize_per_channel_1 = _scale_1 = _zero_point_1 = None 2025-09-09T15:14:02.6064848Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T15:14:02.6065932Z conv2d_5 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_channel_1 = conv1_weight_bias = None 2025-09-09T15:14:02.6067520Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_5, 0.01743602752685547, -29, -128, 127, torch.int8); conv2d_5 = None 2025-09-09T15:14:02.6069117Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01743602752685547, -29, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:14:02.6070230Z quantize_per_channel = self._frozen_param1 2025-09-09T15:14:02.6071334Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:14:02.6073079Z conv2d_4 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_1, dequantize_per_channel, conv2_bias); dequantize_per_tensor_default_1 = dequantize_per_channel = conv2_bias = None 2025-09-09T15:14:02.6075023Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_4, 0.011033773422241211, -1, -128, 127, torch.int8); conv2d_4 = None 2025-09-09T15:14:02.6076933Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011033773422241211, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:14:02.6078200Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:14:02.6078713Z 2025-09-09T15:14:02.6079054Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:14:02.6079505Z onverted model fx: GraphModule( 2025-09-09T15:14:02.6079976Z (conv1): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:14:02.6080611Z (conv2): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:14:02.6081095Z ) 2025-09-09T15:14:02.6081211Z 2025-09-09T15:14:02.6081216Z 2025-09-09T15:14:02.6081220Z 2025-09-09T15:14:02.6081323Z def forward(self, x): 2025-09-09T15:14:02.6082090Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:14:02.6083661Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:14:02.6084920Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:14:02.6086134Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.01743602752685547, -29, -128, 127, torch.int8); conv1 = None 2025-09-09T15:14:02.6087419Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01743602752685547, -29, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:14:02.6088442Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:14:02.6089309Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.011033773422241211, -1, -128, 127, torch.int8); conv2 = None 2025-09-09T15:14:02.6090575Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011033773422241211, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:14:02.6091437Z return dequantize_per_tensor_default_2 2025-09-09T15:14:02.6091710Z 2025-09-09T15:14:02.6091981Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:14:02.6092340Z diff: tensor([[[[0.]], 2025-09-09T15:14:02.6092478Z 2025-09-09T15:14:02.6092553Z [[0.]], 2025-09-09T15:14:02.6092682Z 2025-09-09T15:14:02.6092756Z [[0.]]], 2025-09-09T15:14:02.6092876Z 2025-09-09T15:14:02.6092880Z 2025-09-09T15:14:02.6092957Z [[[0.]], 2025-09-09T15:14:02.6093072Z 2025-09-09T15:14:02.6093145Z [[0.]], 2025-09-09T15:14:02.6093260Z 2025-09-09T15:14:02.6093338Z [[0.]]], 2025-09-09T15:14:02.6093455Z 2025-09-09T15:14:02.6093463Z 2025-09-09T15:14:02.6093534Z [[[0.]], 2025-09-09T15:14:02.6093657Z 2025-09-09T15:14:02.6093727Z [[0.]], 2025-09-09T15:14:02.6093841Z 2025-09-09T15:14:02.6093918Z [[0.]]]]) 2025-09-09T15:14:02.6094124Z model pt2e: GraphModule( 2025-09-09T15:14:02.6094360Z (conv1): Module() 2025-09-09T15:14:02.6094555Z (bn1): Module() 2025-09-09T15:14:02.6094753Z (conv2): Module() 2025-09-09T15:14:02.6094953Z (bn2): Module() 2025-09-09T15:14:02.6095256Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:02.6096192Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:14:02.6097402Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:14:02.6097915Z ) 2025-09-09T15:14:02.6098270Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:02.6099210Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:14:02.6100298Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1921343356370926, max_val=0.1768510341644287) 2025-09-09T15:14:02.6100812Z ) 2025-09-09T15:14:02.6101089Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:02.6102018Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:14:02.6103124Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212442636489868, max_val=0.18097376823425293) 2025-09-09T15:14:02.6103639Z ) 2025-09-09T15:14:02.6103913Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:02.6104842Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:14:02.6105921Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7288155555725098, max_val=2.7138354778289795) 2025-09-09T15:14:02.6106428Z ) 2025-09-09T15:14:02.6106700Z (activation_post_process_4): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:02.6107628Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:14:02.6108711Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4025696516036987, max_val=1.4086220264434814) 2025-09-09T15:14:02.6109213Z ) 2025-09-09T15:14:02.6109383Z ) 2025-09-09T15:14:02.6109478Z 2025-09-09T15:14:02.6109482Z 2025-09-09T15:14:02.6109486Z 2025-09-09T15:14:02.6109568Z def forward(self, x): 2025-09-09T15:14:02.6109857Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:14:02.6110192Z conv1_weight = self.conv1.weight 2025-09-09T15:14:02.6110479Z bn1_weight = self.bn1.weight 2025-09-09T15:14:02.6110738Z bn1_bias = self.bn1.bias 2025-09-09T15:14:02.6110985Z conv2_weight = self.conv2.weight 2025-09-09T15:14:02.6111263Z conv2_bias = self.conv2.bias 2025-09-09T15:14:02.6111526Z bn2_weight = self.bn2.weight 2025-09-09T15:14:02.6111779Z bn2_bias = self.bn2.bias 2025-09-09T15:14:02.6112037Z bn1_running_mean = self.bn1.running_mean 2025-09-09T15:14:02.6112345Z bn1_running_var = self.bn1.running_var 2025-09-09T15:14:02.6112676Z bn1_num_batches_tracked = self.bn1.num_batches_tracked 2025-09-09T15:14:02.6113026Z bn2_running_mean = self.bn2.running_mean 2025-09-09T15:14:02.6113323Z bn2_running_var = self.bn2.running_var 2025-09-09T15:14:02.6113654Z bn2_num_batches_tracked = self.bn2.num_batches_tracked 2025-09-09T15:14:02.6114096Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:14:02.6114684Z add_ = torch.ops.aten.add_.Tensor(bn1_num_batches_tracked, 1); bn1_num_batches_tracked = add_ = None 2025-09-09T15:14:02.6115343Z add__1 = torch.ops.aten.add_.Tensor(bn2_num_batches_tracked, 1); bn2_num_batches_tracked = add__1 = None 2025-09-09T15:14:02.6115869Z add = torch.ops.aten.add.Tensor(bn2_running_var, 1e-05) 2025-09-09T15:14:02.6116515Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:14:02.6116940Z div = torch.ops.aten.div.Tensor(bn2_weight, sqrt); sqrt = None 2025-09-09T15:14:02.6117378Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:14:02.6117981Z mul = torch.ops.aten.mul.Tensor(conv2_weight, reshape); conv2_weight = reshape = None 2025-09-09T15:14:02.6118545Z activation_post_process_3 = self.activation_post_process_3(mul); mul = None 2025-09-09T15:14:20.7519179Z zeros_like = torch.ops.aten.zeros_like.default(conv2_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:14:20.7519787Z add_2 = torch.ops.aten.add.Tensor(bn1_running_var, 1e-05) 2025-09-09T15:14:20.7520278Z sqrt_1 = torch.ops.aten.sqrt.default(add_2); add_2 = None 2025-09-09T15:14:20.7520781Z div_2 = torch.ops.aten.div.Tensor(bn1_weight, sqrt_1); sqrt_1 = None 2025-09-09T15:14:20.7521317Z reshape_3 = torch.ops.aten.reshape.default(div_2, [-1, 1, 1, 1]) 2025-09-09T15:14:20.7521885Z mul_1 = torch.ops.aten.mul.Tensor(conv1_weight, reshape_3); conv1_weight = reshape_3 = None 2025-09-09T15:14:20.7522465Z activation_post_process_1 = self.activation_post_process_1(mul_1); mul_1 = None 2025-09-09T15:14:20.7523315Z conv2d_3 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:14:20.7524150Z reshape_4 = torch.ops.aten.reshape.default(div_2, [1, -1, 1, 1]); div_2 = None 2025-09-09T15:14:20.7524689Z div_3 = torch.ops.aten.div.Tensor(conv2d_3, reshape_4); conv2d_3 = reshape_4 = None 2025-09-09T15:14:20.7525604Z batch_norm_3 = torch.ops.aten.batch_norm.default(div_3, bn1_weight, bn1_bias, bn1_running_mean, bn1_running_var, True, 0.1, 1e-05, True); div_3 = bn1_weight = bn1_bias = bn1_running_mean = bn1_running_var = None 2025-09-09T15:14:20.7526673Z activation_post_process_2 = self.activation_post_process_2(batch_norm_3); batch_norm_3 = None 2025-09-09T15:14:20.7527789Z conv2d_2 = torch.ops.aten.conv2d.default(activation_post_process_2, activation_post_process_3, zeros_like); activation_post_process_2 = activation_post_process_3 = zeros_like = None 2025-09-09T15:14:20.7528669Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:14:20.7529210Z div_1 = torch.ops.aten.div.Tensor(conv2d_2, reshape_1); conv2d_2 = reshape_1 = None 2025-09-09T15:14:20.7529800Z reshape_2 = torch.ops.aten.reshape.default(conv2_bias, [1, -1, 1, 1]); conv2_bias = None 2025-09-09T15:14:20.7530361Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:14:20.7531237Z batch_norm_2 = torch.ops.aten.batch_norm.default(add_1, bn2_weight, bn2_bias, bn2_running_mean, bn2_running_var, True, 0.1, 1e-05, True); add_1 = bn2_weight = bn2_bias = bn2_running_mean = bn2_running_var = None 2025-09-09T15:14:20.7532185Z activation_post_process_4 = self.activation_post_process_4(batch_norm_2); batch_norm_2 = None 2025-09-09T15:14:20.7532774Z return pytree.tree_unflatten((activation_post_process_4,), self._out_spec) 2025-09-09T15:14:20.7533160Z 2025-09-09T15:14:20.7533439Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:14:20.7533798Z model fx: GraphModule( 2025-09-09T15:14:20.7534124Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:20.7535138Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0189]), zero_point=tensor([-17], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:14:20.7536497Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.0985729694366455, max_val=2.7226178646087646) 2025-09-09T15:14:20.7537008Z ) 2025-09-09T15:14:20.7537190Z (conv1): ConvBn2d( 2025-09-09T15:14:20.7537451Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:14:20.7538250Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:14:20.7538718Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:20.7539792Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:14:20.7540906Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212442636489868, max_val=0.18097376823425293) 2025-09-09T15:14:20.7541421Z ) 2025-09-09T15:14:20.7541590Z ) 2025-09-09T15:14:20.7541869Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:20.7542794Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0174]), zero_point=tensor([-29], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:14:20.7543892Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7288155555725098, max_val=2.7138354778289795) 2025-09-09T15:14:20.7544391Z ) 2025-09-09T15:14:20.7544572Z (conv2): ConvBn2d( 2025-09-09T15:14:20.7544811Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:14:20.7545215Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:14:20.7545676Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:20.7546573Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:14:20.7547672Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1921343356370926, max_val=0.1768510341644287) 2025-09-09T15:14:20.7548205Z ) 2025-09-09T15:14:20.7548371Z ) 2025-09-09T15:14:20.7548656Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:14:20.7549576Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0110]), zero_point=tensor([-1], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:14:20.7550668Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.4025696516036987, max_val=1.4086220264434814) 2025-09-09T15:14:20.7551174Z ) 2025-09-09T15:14:20.7551340Z ) 2025-09-09T15:14:20.7551440Z 2025-09-09T15:14:20.7551445Z 2025-09-09T15:14:20.7551448Z 2025-09-09T15:14:20.7551533Z def forward(self, x): 2025-09-09T15:14:20.7551879Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:14:20.7552419Z conv1 = self.conv1(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:14:20.7552972Z activation_post_process_1 = self.activation_post_process_1(conv1); conv1 = None 2025-09-09T15:14:20.7553524Z conv2 = self.conv2(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:14:20.7554071Z activation_post_process_2 = self.activation_post_process_2(conv2); conv2 = None 2025-09-09T15:14:20.7554491Z return activation_post_process_2 2025-09-09T15:14:20.7554749Z 2025-09-09T15:14:20.7555023Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:14:20.7555384Z diff: tensor([[[[0.]], 2025-09-09T15:14:20.7555522Z 2025-09-09T15:14:20.7555602Z [[0.]], 2025-09-09T15:14:20.7555724Z 2025-09-09T15:14:20.7555801Z [[0.]]], 2025-09-09T15:14:20.7555921Z 2025-09-09T15:14:20.7555931Z 2025-09-09T15:14:20.7556005Z [[[0.]], 2025-09-09T15:14:20.7556120Z 2025-09-09T15:14:20.7556310Z [[0.]], 2025-09-09T15:14:20.7556459Z 2025-09-09T15:14:20.7556533Z [[0.]]], 2025-09-09T15:14:20.7556653Z 2025-09-09T15:14:20.7556657Z 2025-09-09T15:14:20.7556732Z [[[0.]], 2025-09-09T15:14:20.7556947Z 2025-09-09T15:14:20.7557019Z [[0.]], 2025-09-09T15:14:20.7557137Z 2025-09-09T15:14:20.7557247Z [[0.]]]], grad_fn=) 2025-09-09T15:14:20.7557541Z converted model pt2e: GraphModule( 2025-09-09T15:14:20.7557805Z (conv1): Module() 2025-09-09T15:14:20.7558003Z (bn1): Module() 2025-09-09T15:14:20.7558209Z (conv2): Module() 2025-09-09T15:14:20.7558488Z (bn2): Module() 2025-09-09T15:14:20.7558687Z ) 2025-09-09T15:14:20.7558787Z 2025-09-09T15:14:20.7558791Z 2025-09-09T15:14:20.7558794Z 2025-09-09T15:14:20.7558883Z def forward(self, x): 2025-09-09T15:14:20.7559160Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:14:20.7559496Z conv2_bias = self.conv2.bias 2025-09-09T15:14:20.7560141Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:14:20.7561376Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:14:20.7562263Z quantize_per_tensor_1 = self._frozen_param0 2025-09-09T15:14:20.7563066Z dequantize_per_tensor_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_1, 0.001512790797278285, 0, -127, 127, torch.int8); quantize_per_tensor_1 = None 2025-09-09T15:14:20.7564231Z conv1_weight_bias = self.conv1.weight_bias 2025-09-09T15:14:20.7565075Z conv2d_5 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor_1, conv1_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor_1 = conv1_weight_bias = None 2025-09-09T15:14:20.7566326Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_5, 0.017422160133719444, -29, -128, 127, torch.int8); conv2d_5 = None 2025-09-09T15:14:20.7567606Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.017422160133719444, -29, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T15:14:20.7568483Z quantize_per_tensor = self._frozen_param1 2025-09-09T15:14:20.7569268Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.001512868795543909, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:14:20.7570524Z conv2d_4 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_3, dequantize_per_tensor, conv2_bias); dequantize_per_tensor_default_3 = dequantize_per_tensor = conv2_bias = None 2025-09-09T15:14:20.7571718Z quantize_per_tensor_default_4 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_4, 0.011024280451238155, -1, -128, 127, torch.int8); conv2d_4 = None 2025-09-09T15:14:20.7572991Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_4, 0.011024280451238155, -1, -128, 127, torch.int8); quantize_per_tensor_default_4 = None 2025-09-09T15:14:20.7573989Z return pytree.tree_unflatten((dequantize_per_tensor_default_4,), self._out_spec) 2025-09-09T15:14:20.7574395Z 2025-09-09T15:14:20.7574677Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:14:20.7575047Z onverted model fx: GraphModule( 2025-09-09T15:14:20.7575429Z (conv1): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:14:20.7575938Z (conv2): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:14:20.7576318Z ) 2025-09-09T15:14:20.7576415Z 2025-09-09T15:14:20.7576419Z 2025-09-09T15:14:20.7576423Z 2025-09-09T15:14:20.7576517Z def forward(self, x): 2025-09-09T15:15:04.1018293Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01890663057565689, -17, -128, 127, torch.int8); x = None 2025-09-09T15:15:04.1019906Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01890663057565689, -17, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:04.1021548Z conv1 = self.conv1(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:15:04.1022807Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv1, 0.017422160133719444, -29, -128, 127, torch.int8); conv1 = None 2025-09-09T15:15:04.1024433Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.017422160133719444, -29, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:15:04.1025738Z conv2 = self.conv2(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:15:04.1026827Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2, 0.011024280451238155, -1, -128, 127, torch.int8); conv2 = None 2025-09-09T15:15:04.1028450Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.011024280451238155, -1, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:15:04.1029559Z return dequantize_per_tensor_default_2 2025-09-09T15:15:04.1029886Z 2025-09-09T15:15:04.1030227Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:04.1030663Z diff: tensor([[[[0.]], 2025-09-09T15:15:04.1030837Z 2025-09-09T15:15:04.1030925Z [[0.]], 2025-09-09T15:15:04.1031067Z 2025-09-09T15:15:04.1031160Z [[0.]]], 2025-09-09T15:15:04.1031304Z 2025-09-09T15:15:04.1031309Z 2025-09-09T15:15:04.1031394Z [[[0.]], 2025-09-09T15:15:04.1031535Z 2025-09-09T15:15:04.1031632Z [[0.]], 2025-09-09T15:15:04.1031768Z 2025-09-09T15:15:04.1031854Z [[0.]]], 2025-09-09T15:15:04.1032002Z 2025-09-09T15:15:04.1032014Z 2025-09-09T15:15:04.1032102Z [[[0.]], 2025-09-09T15:15:04.1032239Z 2025-09-09T15:15:04.1032331Z [[0.]], 2025-09-09T15:15:04.1032468Z 2025-09-09T15:15:04.1032553Z [[0.]]]]) 2025-09-09T15:15:04.1033030Z PASSED 2025-09-09T15:15:04.1033894Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_per_channel_weight_bias PASSED 2025-09-09T15:15:04.1035134Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_relu_fusion model pt2e: GraphModule( 2025-09-09T15:15:04.1035892Z (conv): Module() 2025-09-09T15:15:04.1036130Z (bn): Module() 2025-09-09T15:15:04.1036571Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:04.1037720Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:04.1039072Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:04.1039681Z ) 2025-09-09T15:15:04.1040010Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:04.1041212Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0014, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:15:04.1042782Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1835, -0.1822, -0.1883]), max_val=tensor([0.1799, 0.1856, 0.1719])) 2025-09-09T15:15:04.1043574Z ) 2025-09-09T15:15:04.1043890Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:04.1045038Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:04.1046476Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2505061626434326) 2025-09-09T15:15:04.1047042Z ) 2025-09-09T15:15:04.1047245Z ) 2025-09-09T15:15:04.1047358Z 2025-09-09T15:15:04.1047444Z 2025-09-09T15:15:04.1047450Z 2025-09-09T15:15:04.1047550Z def forward(self, x): 2025-09-09T15:15:04.1047891Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:04.1048292Z conv_weight = self.conv.weight 2025-09-09T15:15:04.1048608Z conv_bias = self.conv.bias 2025-09-09T15:15:04.1048910Z bn_weight = self.bn.weight 2025-09-09T15:15:04.1049200Z bn_bias = self.bn.bias 2025-09-09T15:15:04.1049508Z bn_running_mean = self.bn.running_mean 2025-09-09T15:15:04.1049863Z bn_running_var = self.bn.running_var 2025-09-09T15:15:04.1050265Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:15:04.1050800Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:04.1051519Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:15:04.1052162Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:15:04.1052628Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:15:04.1053121Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:15:04.1053647Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:15:04.1054287Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:15:04.1054971Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:15:04.1055719Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:15:04.1056936Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:15:04.1058046Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:15:04.1058701Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:15:04.1059406Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:15:04.1060078Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:15:04.1061130Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:15:04.1062174Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:15:04.1062801Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:15:04.1063475Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:15:04.1064290Z 2025-09-09T15:15:04.1064630Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:04.1065068Z model fx: GraphModule( 2025-09-09T15:15:04.1065457Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:04.1066606Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:04.1067942Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:04.1068558Z ) 2025-09-09T15:15:04.1068770Z (conv): ConvBnReLU2d( 2025-09-09T15:15:04.1069059Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:15:04.1069548Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:15:04.1070307Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:04.1071585Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0014, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:15:04.1073248Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1835, -0.1822, -0.1883]), max_val=tensor([0.1799, 0.1856, 0.1719])) 2025-09-09T15:15:04.1074060Z ) 2025-09-09T15:15:04.1074231Z ) 2025-09-09T15:15:04.1074510Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:04.1075452Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:04.1076583Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2505061626434326) 2025-09-09T15:15:04.1077052Z ) 2025-09-09T15:15:04.1077218Z ) 2025-09-09T15:15:04.1077324Z 2025-09-09T15:15:04.1077329Z 2025-09-09T15:15:04.1077333Z 2025-09-09T15:15:04.1077426Z def forward(self, x): 2025-09-09T15:15:04.1077780Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:04.1078313Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:15:04.1078859Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:15:04.1079280Z return activation_post_process_1 2025-09-09T15:15:04.1079542Z 2025-09-09T15:15:04.1079817Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:04.1080186Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:15:04.1080426Z [0., 0., 0.], 2025-09-09T15:15:04.1080644Z [0., 0., 0.]], 2025-09-09T15:15:04.1080783Z 2025-09-09T15:15:04.1080866Z [[0., 0., 0.], 2025-09-09T15:15:04.1081069Z [0., 0., 0.], 2025-09-09T15:15:04.1081282Z [0., 0., 0.]], 2025-09-09T15:15:04.1081419Z 2025-09-09T15:15:04.1081494Z [[0., 0., 0.], 2025-09-09T15:15:04.1081706Z [0., 0., 0.], 2025-09-09T15:15:04.1081941Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:15:04.1082256Z converted model pt2e: GraphModule( 2025-09-09T15:15:04.1082523Z (conv): Module() 2025-09-09T15:15:04.1082731Z (bn): Module() 2025-09-09T15:15:04.1082926Z ) 2025-09-09T15:15:04.1083024Z 2025-09-09T15:15:04.1083028Z 2025-09-09T15:15:04.1083032Z 2025-09-09T15:15:04.1083119Z def forward(self, x): 2025-09-09T15:15:04.1083404Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:04.1083732Z conv_bias = self.conv.bias 2025-09-09T15:15:04.1084377Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:15:04.1085620Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:04.1086469Z _scale_0 = self._scale_0 2025-09-09T15:15:04.1086732Z _zero_point_0 = self._zero_point_0 2025-09-09T15:15:04.1087033Z quantize_per_channel = self._frozen_param0 2025-09-09T15:15:21.9216398Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:15:21.9218179Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:15:21.9219538Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T15:15:21.9220528Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.004903945606201887, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:15:21.9222339Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.004903945606201887, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:15:21.9223619Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:15:21.9224141Z 2025-09-09T15:15:21.9224483Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:21.9224955Z onverted model fx: GraphModule( 2025-09-09T15:15:21.9225283Z (conv): ConvReLU2d( 2025-09-09T15:15:21.9225695Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:15:21.9226183Z (1): ReLU() 2025-09-09T15:15:21.9226417Z ) 2025-09-09T15:15:21.9226665Z ) 2025-09-09T15:15:21.9226806Z 2025-09-09T15:15:21.9226813Z 2025-09-09T15:15:21.9226819Z 2025-09-09T15:15:21.9226925Z def forward(self, x): 2025-09-09T15:15:21.9227709Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:15:21.9229283Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:21.9230553Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:15:21.9231644Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.004903945606201887, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:15:21.9233282Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.004903945606201887, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:15:21.9234403Z return dequantize_per_tensor_default_1 2025-09-09T15:15:21.9234770Z 2025-09-09T15:15:21.9235111Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:21.9235571Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:15:21.9235852Z [0., 0., 0.], 2025-09-09T15:15:21.9236103Z [0., 0., 0.]], 2025-09-09T15:15:21.9236337Z 2025-09-09T15:15:21.9236439Z [[0., 0., 0.], 2025-09-09T15:15:21.9236686Z [0., 0., 0.], 2025-09-09T15:15:21.9236938Z [0., 0., 0.]], 2025-09-09T15:15:21.9237106Z 2025-09-09T15:15:21.9237198Z [[0., 0., 0.], 2025-09-09T15:15:21.9237444Z [0., 0., 0.], 2025-09-09T15:15:21.9237688Z [0., 0., 0.]]]]) 2025-09-09T15:15:21.9237972Z model pt2e: GraphModule( 2025-09-09T15:15:21.9238242Z (conv): Module() 2025-09-09T15:15:21.9238491Z (bn): Module() 2025-09-09T15:15:21.9238842Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:21.9240018Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:21.9241399Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:21.9242031Z ) 2025-09-09T15:15:21.9242362Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:21.9243529Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:15:21.9244889Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1882954239845276, max_val=0.1855725795030594) 2025-09-09T15:15:21.9245614Z ) 2025-09-09T15:15:21.9245932Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:21.9247152Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:21.9248469Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2396948337554932) 2025-09-09T15:15:21.9249050Z ) 2025-09-09T15:15:21.9249254Z ) 2025-09-09T15:15:21.9249367Z 2025-09-09T15:15:21.9249372Z 2025-09-09T15:15:21.9249377Z 2025-09-09T15:15:21.9249477Z def forward(self, x): 2025-09-09T15:15:21.9249812Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:21.9250231Z conv_weight = self.conv.weight 2025-09-09T15:15:21.9250555Z conv_bias = self.conv.bias 2025-09-09T15:15:21.9250871Z bn_weight = self.bn.weight 2025-09-09T15:15:21.9251169Z bn_bias = self.bn.bias 2025-09-09T15:15:21.9251475Z bn_running_mean = self.bn.running_mean 2025-09-09T15:15:21.9251838Z bn_running_var = self.bn.running_var 2025-09-09T15:15:21.9252231Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:15:21.9252767Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:21.9253471Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:15:21.9254108Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:15:21.9254565Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:15:21.9255057Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:15:21.9255589Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:15:21.9256199Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:15:21.9256891Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:15:21.9257631Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:15:21.9258824Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:15:21.9259906Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:15:21.9260549Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:15:21.9261255Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:15:21.9261923Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:15:21.9262973Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:15:21.9264330Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:15:21.9264961Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:15:21.9265618Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:15:21.9266083Z 2025-09-09T15:15:21.9266416Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:21.9266848Z model fx: GraphModule( 2025-09-09T15:15:21.9267229Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:21.9268400Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:21.9269910Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:21.9270593Z ) 2025-09-09T15:15:21.9270818Z (conv): ConvBnReLU2d( 2025-09-09T15:15:21.9271108Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:15:21.9271640Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:15:21.9272102Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:21.9273011Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:15:21.9274108Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.1882954239845276, max_val=0.1855725795030594) 2025-09-09T15:15:21.9274617Z ) 2025-09-09T15:15:21.9274787Z ) 2025-09-09T15:15:21.9275062Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:21.9276002Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0049]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:21.9277130Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.2396948337554932) 2025-09-09T15:15:21.9277598Z ) 2025-09-09T15:15:21.9277757Z ) 2025-09-09T15:15:21.9277862Z 2025-09-09T15:15:21.9277866Z 2025-09-09T15:15:21.9277870Z 2025-09-09T15:15:21.9277952Z def forward(self, x): 2025-09-09T15:15:21.9278295Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:21.9278830Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:15:21.9279374Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:15:21.9279793Z return activation_post_process_1 2025-09-09T15:15:21.9280056Z 2025-09-09T15:15:21.9280331Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:21.9280698Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:15:21.9280933Z [0., 0., 0.], 2025-09-09T15:15:21.9281145Z [0., 0., 0.]], 2025-09-09T15:15:21.9281282Z 2025-09-09T15:15:21.9281372Z [[0., 0., 0.], 2025-09-09T15:15:21.9281573Z [0., 0., 0.], 2025-09-09T15:15:21.9281783Z [0., 0., 0.]], 2025-09-09T15:15:21.9281919Z 2025-09-09T15:15:21.9281993Z [[0., 0., 0.], 2025-09-09T15:15:21.9282200Z [0., 0., 0.], 2025-09-09T15:15:21.9282435Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:15:21.9282744Z converted model pt2e: GraphModule( 2025-09-09T15:15:21.9283004Z (conv): Module() 2025-09-09T15:15:21.9283208Z (bn): Module() 2025-09-09T15:15:21.9283394Z ) 2025-09-09T15:15:21.9283503Z 2025-09-09T15:15:21.9283507Z 2025-09-09T15:15:21.9283511Z 2025-09-09T15:15:21.9283600Z def forward(self, x): 2025-09-09T15:15:21.9283882Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:21.9284208Z conv_bias = self.conv.bias 2025-09-09T15:15:33.8421969Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:15:33.8424427Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:33.8425307Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:15:33.8426114Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014826410915702581, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:15:33.8427361Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:15:33.8428629Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T15:15:33.8429556Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.004861548542976379, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:15:33.8430825Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.004861548542976379, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:15:33.8431838Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:15:33.8432243Z 2025-09-09T15:15:33.8432531Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:33.8432900Z onverted model fx: GraphModule( 2025-09-09T15:15:33.8433162Z (conv): ConvReLU2d( 2025-09-09T15:15:33.8433519Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:15:33.8433883Z (1): ReLU() 2025-09-09T15:15:33.8434076Z ) 2025-09-09T15:15:33.8434242Z ) 2025-09-09T15:15:33.8434342Z 2025-09-09T15:15:33.8434352Z 2025-09-09T15:15:33.8434356Z 2025-09-09T15:15:33.8434439Z def forward(self, x): 2025-09-09T15:15:33.8435057Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:15:33.8436359Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:33.8437358Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:15:33.8438203Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.004861548542976379, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:15:33.8439468Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.004861548542976379, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:15:33.8440341Z return dequantize_per_tensor_default_1 2025-09-09T15:15:33.8440613Z 2025-09-09T15:15:33.8440891Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:33.8441249Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:15:33.8441484Z [0., 0., 0.], 2025-09-09T15:15:33.8441685Z [0., 0., 0.]], 2025-09-09T15:15:33.8441827Z 2025-09-09T15:15:33.8441901Z [[0., 0., 0.], 2025-09-09T15:15:33.8442099Z [0., 0., 0.], 2025-09-09T15:15:33.8442309Z [0., 0., 0.]], 2025-09-09T15:15:33.8442443Z 2025-09-09T15:15:33.8442520Z [[0., 0., 0.], 2025-09-09T15:15:33.8442718Z [0., 0., 0.], 2025-09-09T15:15:33.8442926Z [0., 0., 0.]]]]) 2025-09-09T15:15:33.8443360Z PASSED 2025-09-09T15:15:33.8443962Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_relu_fusion_cuda model pt2e: GraphModule( 2025-09-09T15:15:33.8444583Z (conv): Module() 2025-09-09T15:15:33.8444785Z (bn): Module() 2025-09-09T15:15:33.8445083Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:33.8446192Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:33.8447448Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:33.8447948Z ) 2025-09-09T15:15:33.8448226Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:33.8449471Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:15:33.8451071Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T15:15:33.8451798Z ) 2025-09-09T15:15:33.8452073Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:33.8453187Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0081], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:33.8454421Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0564441680908203) 2025-09-09T15:15:33.8454888Z ) 2025-09-09T15:15:33.8455058Z ) 2025-09-09T15:15:33.8455153Z 2025-09-09T15:15:33.8455157Z 2025-09-09T15:15:33.8455161Z 2025-09-09T15:15:33.8455242Z def forward(self, x): 2025-09-09T15:15:33.8455536Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:33.8455874Z conv_weight = self.conv.weight 2025-09-09T15:15:33.8456141Z conv_bias = self.conv.bias 2025-09-09T15:15:33.8456396Z bn_weight = self.bn.weight 2025-09-09T15:15:33.8456640Z bn_bias = self.bn.bias 2025-09-09T15:15:33.8456900Z bn_running_mean = self.bn.running_mean 2025-09-09T15:15:33.8457190Z bn_running_var = self.bn.running_var 2025-09-09T15:15:33.8457521Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:15:33.8457955Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:33.8458550Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:15:33.8459079Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:15:33.8459465Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:15:33.8459885Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:15:33.8460324Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:15:33.8460835Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:15:33.8461389Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:15:33.8462002Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:15:33.8462972Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:15:33.8464190Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:15:33.8464779Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:15:33.8465368Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:15:33.8465928Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:15:33.8466800Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:15:33.8467640Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:15:33.8468159Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:15:33.8468698Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:15:33.8469207Z 2025-09-09T15:15:33.8469484Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:33.8469838Z model fx: GraphModule( 2025-09-09T15:15:33.8470160Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:33.8471378Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:33.8472637Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:33.8473139Z ) 2025-09-09T15:15:33.8473317Z (conv): ConvBnReLU2d( 2025-09-09T15:15:33.8473559Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:15:33.8473959Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:15:33.8474438Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:33.8475601Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015, 0.0015, 0.0014], device='cuda:0'), zero_point=tensor([0, 0, 0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:15:33.8477222Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787], device='cuda:0'), max_val=tensor([0.1824, 0.1870, 0.1478], device='cuda:0')) 2025-09-09T15:15:33.8477948Z ) 2025-09-09T15:15:33.8478121Z ) 2025-09-09T15:15:33.8478400Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:51.8524591Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0081], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:51.8526678Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0564441680908203) 2025-09-09T15:15:51.8527144Z ) 2025-09-09T15:15:51.8527329Z ) 2025-09-09T15:15:51.8527444Z 2025-09-09T15:15:51.8527449Z 2025-09-09T15:15:51.8527453Z 2025-09-09T15:15:51.8527541Z def forward(self, x): 2025-09-09T15:15:51.8527904Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:51.8528448Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:15:51.8528992Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:15:51.8529423Z return activation_post_process_1 2025-09-09T15:15:51.8529681Z 2025-09-09T15:15:51.8529970Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:51.8530341Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:15:51.8530583Z [0., 0., 0.], 2025-09-09T15:15:51.8530789Z [0., 0., 0.]], 2025-09-09T15:15:51.8530937Z 2025-09-09T15:15:51.8531030Z [[0., 0., 0.], 2025-09-09T15:15:51.8531241Z [0., 0., 0.], 2025-09-09T15:15:51.8531440Z [0., 0., 0.]], 2025-09-09T15:15:51.8531582Z 2025-09-09T15:15:51.8541101Z [[0., 0., 0.], 2025-09-09T15:15:51.8541334Z [0., 0., 0.], 2025-09-09T15:15:51.8541607Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T15:15:51.8541933Z converted model pt2e: GraphModule( 2025-09-09T15:15:51.8542197Z (conv): Module() 2025-09-09T15:15:51.8542398Z (bn): Module() 2025-09-09T15:15:51.8542593Z ) 2025-09-09T15:15:51.8542688Z 2025-09-09T15:15:51.8542692Z 2025-09-09T15:15:51.8542696Z 2025-09-09T15:15:51.8542789Z def forward(self, x): 2025-09-09T15:15:51.8543067Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:51.8543622Z conv_bias = self.conv.bias 2025-09-09T15:15:51.8544256Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:15:51.8545609Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:51.8546459Z _scale_0 = self._scale_0 2025-09-09T15:15:51.8546710Z _zero_point_0 = self._zero_point_0 2025-09-09T15:15:51.8547012Z quantize_per_channel = self._frozen_param0 2025-09-09T15:15:51.8547878Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:15:51.8549216Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:15:51.8550063Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T15:15:51.8550834Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.008064487017691135, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:15:51.8552101Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008064487017691135, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:15:51.8553096Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:15:51.8553504Z 2025-09-09T15:15:51.8553775Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:51.8554146Z onverted model fx: GraphModule( 2025-09-09T15:15:51.8554395Z (conv): ConvReLU2d( 2025-09-09T15:15:51.8554727Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:15:51.8555095Z (1): ReLU() 2025-09-09T15:15:51.8555277Z ) 2025-09-09T15:15:51.8555443Z ) 2025-09-09T15:15:51.8555535Z 2025-09-09T15:15:51.8555539Z 2025-09-09T15:15:51.8555543Z 2025-09-09T15:15:51.8555626Z def forward(self, x): 2025-09-09T15:15:51.8556331Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:15:51.8557570Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:15:51.8558561Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:15:51.8559407Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.008064487017691135, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:15:51.8560675Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.008064487017691135, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:15:51.8561544Z return dequantize_per_tensor_default_1 2025-09-09T15:15:51.8561816Z 2025-09-09T15:15:51.8562085Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:51.8562451Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:15:51.8562682Z [0., 0., 0.], 2025-09-09T15:15:51.8562887Z [0., 0., 0.]], 2025-09-09T15:15:51.8563021Z 2025-09-09T15:15:51.8563097Z [[0., 0., 0.], 2025-09-09T15:15:51.8563296Z [0., 0., 0.], 2025-09-09T15:15:51.8563497Z [0., 0., 0.]], 2025-09-09T15:15:51.8563632Z 2025-09-09T15:15:51.8564018Z [[0., 0., 0.], 2025-09-09T15:15:51.8564224Z [0., 0., 0.], 2025-09-09T15:15:51.8564571Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T15:15:51.8564842Z model pt2e: GraphModule( 2025-09-09T15:15:51.8565057Z (conv): Module() 2025-09-09T15:15:51.8565252Z (bn): Module() 2025-09-09T15:15:51.8565539Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:51.8566795Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:51.8568049Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:15:51.8568545Z ) 2025-09-09T15:15:51.8568814Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:51.8569910Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:15:51.8571178Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T15:15:51.8571685Z ) 2025-09-09T15:15:51.8571951Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:51.8573049Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0080], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:15:51.8574258Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0522286891937256) 2025-09-09T15:15:51.8574713Z ) 2025-09-09T15:15:51.8574877Z ) 2025-09-09T15:15:51.8574974Z 2025-09-09T15:15:51.8574978Z 2025-09-09T15:15:51.8574983Z 2025-09-09T15:15:51.8575061Z def forward(self, x): 2025-09-09T15:15:51.8575339Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:15:51.8575660Z conv_weight = self.conv.weight 2025-09-09T15:15:51.8575935Z conv_bias = self.conv.bias 2025-09-09T15:15:51.8576187Z bn_weight = self.bn.weight 2025-09-09T15:15:51.8576426Z bn_bias = self.bn.bias 2025-09-09T15:15:51.8576683Z bn_running_mean = self.bn.running_mean 2025-09-09T15:15:51.8576970Z bn_running_var = self.bn.running_var 2025-09-09T15:15:51.8577292Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:15:51.8577717Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:15:51.8578301Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:15:51.8578819Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:15:51.8579202Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:15:51.8579608Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:15:51.8580041Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:15:51.8580547Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:15:51.8581097Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:15:51.8581703Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:15:51.8582661Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:15:51.8583532Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:15:51.8584067Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:15:51.8584723Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:15:51.8585277Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:15:51.8586211Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:15:51.8587036Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:15:51.8587549Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:15:51.8588076Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:15:51.8588456Z 2025-09-09T15:15:51.8588723Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:15:51.8589077Z model fx: GraphModule( 2025-09-09T15:15:51.8589401Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:15:51.8590500Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0183], device='cuda:0'), zero_point=tensor([10], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:18.8700526Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:18.8701245Z ) 2025-09-09T15:16:18.8701470Z (conv): ConvBnReLU2d( 2025-09-09T15:16:18.8701765Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:16:18.8702277Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:16:18.8702839Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:18.8704290Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0015], device='cuda:0'), zero_point=tensor([0], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:16:18.8705995Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965020775794983, max_val=0.1870359182357788) 2025-09-09T15:16:18.8706642Z ) 2025-09-09T15:16:18.8706852Z ) 2025-09-09T15:16:18.8707185Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:18.8708601Z fake_quant_enabled=tensor([1], device='cuda:0'), observer_enabled=tensor([1], device='cuda:0'), scale=tensor([0.0080], device='cuda:0'), zero_point=tensor([-128], device='cuda:0', dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:18.8710169Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=2.0522286891937256) 2025-09-09T15:16:18.8710759Z ) 2025-09-09T15:16:18.8710969Z ) 2025-09-09T15:16:18.8711085Z 2025-09-09T15:16:18.8711090Z 2025-09-09T15:16:18.8711095Z 2025-09-09T15:16:18.8711196Z def forward(self, x): 2025-09-09T15:16:18.8711620Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:18.8712280Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:16:18.8712945Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:16:18.8713470Z return activation_post_process_1 2025-09-09T15:16:18.8713782Z 2025-09-09T15:16:18.8714116Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:18.8714561Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:18.8714855Z [0., 0., 0.], 2025-09-09T15:16:18.8715102Z [0., 0., 0.]], 2025-09-09T15:16:18.8715276Z 2025-09-09T15:16:18.8715367Z [[0., 0., 0.], 2025-09-09T15:16:18.8715620Z [0., 0., 0.], 2025-09-09T15:16:18.8716467Z [0., 0., 0.]], 2025-09-09T15:16:18.8716636Z 2025-09-09T15:16:18.8716733Z [[0., 0., 0.], 2025-09-09T15:16:18.8716976Z [0., 0., 0.], 2025-09-09T15:16:18.8717296Z [0., 0., 0.]]]], device='cuda:0', grad_fn=) 2025-09-09T15:16:18.8717694Z converted model pt2e: GraphModule( 2025-09-09T15:16:18.8718202Z (conv): Module() 2025-09-09T15:16:18.8718443Z (bn): Module() 2025-09-09T15:16:18.8718673Z ) 2025-09-09T15:16:18.8718789Z 2025-09-09T15:16:18.8718794Z 2025-09-09T15:16:18.8718799Z 2025-09-09T15:16:18.8718904Z def forward(self, x): 2025-09-09T15:16:18.8719239Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:18.8719647Z conv_bias = self.conv.bias 2025-09-09T15:16:18.8720444Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:18.8722004Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:18.8723150Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:16:18.8724135Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014933086931705475, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:16:18.8725411Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:16:18.8726255Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T15:16:18.8727032Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.00804795604199171, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:16:18.8728316Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.00804795604199171, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:16:18.8729334Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:16:18.8729752Z 2025-09-09T15:16:18.8730039Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:18.8730424Z onverted model fx: GraphModule( 2025-09-09T15:16:18.8730684Z (conv): ConvReLU2d( 2025-09-09T15:16:18.8731018Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:16:18.8731398Z (1): ReLU() 2025-09-09T15:16:18.8731585Z ) 2025-09-09T15:16:18.8731757Z ) 2025-09-09T15:16:18.8731853Z 2025-09-09T15:16:18.8731857Z 2025-09-09T15:16:18.8731861Z 2025-09-09T15:16:18.8731951Z def forward(self, x): 2025-09-09T15:16:18.8732573Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:18.8733815Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:18.8734829Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:16:18.8735695Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.00804795604199171, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:16:18.8736968Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.00804795604199171, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:18.8737857Z return dequantize_per_tensor_default_1 2025-09-09T15:16:18.8738139Z 2025-09-09T15:16:18.8738420Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:18.8738896Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:18.8739130Z [0., 0., 0.], 2025-09-09T15:16:18.8739348Z [0., 0., 0.]], 2025-09-09T15:16:18.8739486Z 2025-09-09T15:16:18.8739559Z [[0., 0., 0.], 2025-09-09T15:16:18.8739769Z [0., 0., 0.], 2025-09-09T15:16:18.8740057Z [0., 0., 0.]], 2025-09-09T15:16:18.8740199Z 2025-09-09T15:16:18.8740271Z [[0., 0., 0.], 2025-09-09T15:16:18.8740477Z [0., 0., 0.], 2025-09-09T15:16:18.8740693Z [0., 0., 0.]]]], device='cuda:0') 2025-09-09T15:16:18.8741187Z PASSED 2025-09-09T15:16:18.8741818Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_relu_fusion_no_conv_bias model pt2e: GraphModule( 2025-09-09T15:16:18.8742478Z (conv): Module() 2025-09-09T15:16:18.8742677Z (bn): Module() 2025-09-09T15:16:18.8742979Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:18.8743928Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:18.8745039Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:18.8745548Z ) 2025-09-09T15:16:18.8745820Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:18.8746813Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:16:18.8748100Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1866, -0.1825, -0.1912]), max_val=tensor([0.1747, 0.1914, 0.1702])) 2025-09-09T15:16:18.8748773Z ) 2025-09-09T15:16:18.8749046Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:18.8749995Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:18.8751049Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.984864354133606) 2025-09-09T15:16:18.8751515Z ) 2025-09-09T15:16:18.8751685Z ) 2025-09-09T15:16:18.8751780Z 2025-09-09T15:16:18.8751784Z 2025-09-09T15:16:18.8751788Z 2025-09-09T15:16:18.8751870Z def forward(self, x): 2025-09-09T15:16:18.8752153Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:18.8752483Z conv_weight = self.conv.weight 2025-09-09T15:16:18.8752759Z bn_weight = self.bn.weight 2025-09-09T15:16:18.8753007Z bn_bias = self.bn.bias 2025-09-09T15:16:18.8753268Z bn_running_mean = self.bn.running_mean 2025-09-09T15:16:18.8753573Z bn_running_var = self.bn.running_var 2025-09-09T15:16:18.8753902Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:16:18.8754344Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:18.8754936Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:16:18.8755466Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:16:18.8755852Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:16:18.8756325Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:16:18.8756774Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:16:18.8757278Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:16:18.8757839Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:16:18.8758766Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:16:18.8759595Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:16:18.8760206Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:16:18.8761097Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:16:18.8761940Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:16:18.8762457Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:16:18.8762998Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:16:18.8763380Z 2025-09-09T15:16:18.8763670Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:18.8764333Z model fx: GraphModule( 2025-09-09T15:16:36.6216844Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:36.6218119Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:36.6219526Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:36.6220162Z ) 2025-09-09T15:16:36.6220385Z (conv): ConvBnReLU2d( 2025-09-09T15:16:36.6220704Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:16:36.6221231Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:16:36.6221798Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:36.6223020Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:16:36.6224673Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1866, -0.1825, -0.1912]), max_val=tensor([0.1747, 0.1914, 0.1702])) 2025-09-09T15:16:36.6225497Z ) 2025-09-09T15:16:36.6225699Z ) 2025-09-09T15:16:36.6226032Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:36.6227215Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:36.6228529Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.984864354133606) 2025-09-09T15:16:36.6229109Z ) 2025-09-09T15:16:36.6229313Z ) 2025-09-09T15:16:36.6229429Z 2025-09-09T15:16:36.6229434Z 2025-09-09T15:16:36.6229439Z 2025-09-09T15:16:36.6229550Z def forward(self, x): 2025-09-09T15:16:36.6229968Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:36.6230625Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:16:36.6231294Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:16:36.6231817Z return activation_post_process_1 2025-09-09T15:16:36.6232131Z 2025-09-09T15:16:36.6232459Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:36.6232905Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:36.6233186Z [0., 0., 0.], 2025-09-09T15:16:36.6233444Z [0., 0., 0.]], 2025-09-09T15:16:36.6233612Z 2025-09-09T15:16:36.6233706Z [[0., 0., 0.], 2025-09-09T15:16:36.6233959Z [0., 0., 0.], 2025-09-09T15:16:36.6234634Z [0., 0., 0.]], 2025-09-09T15:16:36.6234809Z 2025-09-09T15:16:36.6234900Z [[0., 0., 0.], 2025-09-09T15:16:36.6235153Z [0., 0., 0.], 2025-09-09T15:16:36.6235433Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:16:36.6235810Z converted model pt2e: GraphModule( 2025-09-09T15:16:36.6236416Z (conv): Module() 2025-09-09T15:16:36.6236670Z (bn): Module() 2025-09-09T15:16:36.6236901Z ) 2025-09-09T15:16:36.6237022Z 2025-09-09T15:16:36.6237027Z 2025-09-09T15:16:36.6237032Z 2025-09-09T15:16:36.6237135Z def forward(self, x): 2025-09-09T15:16:36.6237468Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:36.6238371Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:36.6239951Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:36.6241039Z _scale_0 = self._scale_0 2025-09-09T15:16:36.6241354Z _zero_point_0 = self._zero_point_0 2025-09-09T15:16:36.6241713Z quantize_per_channel = self._frozen_param0 2025-09-09T15:16:36.6242812Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:16:36.6243905Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:16:36.6244941Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_weight_bias = None 2025-09-09T15:16:36.6246058Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T15:16:36.6247020Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.007783781737089157, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:16:36.6248619Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007783781737089157, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:36.6249878Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:16:36.6250379Z 2025-09-09T15:16:36.6250713Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:36.6251165Z onverted model fx: GraphModule( 2025-09-09T15:16:36.6251475Z (conv): ConvReLU2d( 2025-09-09T15:16:36.6251886Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:16:36.6252339Z (1): ReLU() 2025-09-09T15:16:36.6252570Z ) 2025-09-09T15:16:36.6252770Z ) 2025-09-09T15:16:36.6252883Z 2025-09-09T15:16:36.6252895Z 2025-09-09T15:16:36.6252906Z 2025-09-09T15:16:36.6253006Z def forward(self, x): 2025-09-09T15:16:36.6253754Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:36.6255296Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:36.6256549Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:16:36.6257623Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007783781737089157, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:16:36.6259288Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007783781737089157, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:36.6260486Z return dequantize_per_tensor_default_1 2025-09-09T15:16:36.6260759Z 2025-09-09T15:16:36.6261038Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:36.6261405Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:36.6261645Z [0., 0., 0.], 2025-09-09T15:16:36.6261938Z [0., 0., 0.]], 2025-09-09T15:16:36.6262087Z 2025-09-09T15:16:36.6262161Z [[0., 0., 0.], 2025-09-09T15:16:36.6262360Z [0., 0., 0.], 2025-09-09T15:16:36.6262568Z [0., 0., 0.]], 2025-09-09T15:16:36.6262705Z 2025-09-09T15:16:36.6262785Z [[0., 0., 0.], 2025-09-09T15:16:36.6262983Z [0., 0., 0.], 2025-09-09T15:16:36.6263192Z [0., 0., 0.]]]]) 2025-09-09T15:16:36.6263427Z model pt2e: GraphModule( 2025-09-09T15:16:36.6263656Z (conv): Module() 2025-09-09T15:16:36.6264134Z (bn): Module() 2025-09-09T15:16:36.6264439Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:36.6265378Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:36.6266489Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:36.6266996Z ) 2025-09-09T15:16:36.6267268Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:36.6268215Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:16:36.6269332Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19124282896518707, max_val=0.19141820073127747) 2025-09-09T15:16:36.6269841Z ) 2025-09-09T15:16:36.6270119Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:36.6271058Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:36.6272121Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.9838534593582153) 2025-09-09T15:16:36.6272583Z ) 2025-09-09T15:16:36.6272749Z ) 2025-09-09T15:16:36.6272843Z 2025-09-09T15:16:36.6272848Z 2025-09-09T15:16:36.6272851Z 2025-09-09T15:16:36.6272938Z def forward(self, x): 2025-09-09T15:16:36.6273215Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:36.6273555Z conv_weight = self.conv.weight 2025-09-09T15:16:36.6273823Z bn_weight = self.bn.weight 2025-09-09T15:16:36.6274076Z bn_bias = self.bn.bias 2025-09-09T15:16:36.6274328Z bn_running_mean = self.bn.running_mean 2025-09-09T15:16:36.6274632Z bn_running_var = self.bn.running_var 2025-09-09T15:16:36.6274971Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:16:36.6275404Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:36.6275994Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:16:36.6276567Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:16:36.6276959Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:16:36.6277368Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:16:36.6277814Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:16:36.6278323Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:16:36.6278881Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:16:36.6279721Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, None); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:16:36.6280716Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:16:36.6281261Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:16:36.6282264Z batch_norm_1 = torch.ops.aten.batch_norm.default(div_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); div_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:16:36.6283105Z relu = torch.ops.aten.relu.default(batch_norm_1); batch_norm_1 = None 2025-09-09T15:16:36.6283632Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:16:56.1864890Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:16:56.1865461Z 2025-09-09T15:16:56.1865810Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:56.1866300Z model fx: GraphModule( 2025-09-09T15:16:56.1866687Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1867897Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:56.1869317Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:56.1869949Z ) 2025-09-09T15:16:56.1870181Z (conv): ConvBnReLU2d( 2025-09-09T15:16:56.1870500Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:16:56.1871026Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:16:56.1871604Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1872758Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:16:56.1874194Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19124282896518707, max_val=0.19141820073127747) 2025-09-09T15:16:56.1874861Z ) 2025-09-09T15:16:56.1875068Z ) 2025-09-09T15:16:56.1875403Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1876671Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0078]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:56.1878028Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.9838534593582153) 2025-09-09T15:16:56.1878612Z ) 2025-09-09T15:16:56.1878823Z ) 2025-09-09T15:16:56.1878940Z 2025-09-09T15:16:56.1878945Z 2025-09-09T15:16:56.1878957Z 2025-09-09T15:16:56.1879071Z def forward(self, x): 2025-09-09T15:16:56.1879496Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:56.1880160Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:16:56.1880841Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:16:56.1881375Z return activation_post_process_1 2025-09-09T15:16:56.1881703Z 2025-09-09T15:16:56.1882041Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:56.1882493Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:56.1891398Z [0., 0., 0.], 2025-09-09T15:16:56.1891641Z [0., 0., 0.]], 2025-09-09T15:16:56.1891786Z 2025-09-09T15:16:56.1891867Z [[0., 0., 0.], 2025-09-09T15:16:56.1892082Z [0., 0., 0.], 2025-09-09T15:16:56.1892286Z [0., 0., 0.]], 2025-09-09T15:16:56.1892433Z 2025-09-09T15:16:56.1892509Z [[0., 0., 0.], 2025-09-09T15:16:56.1893198Z [0., 0., 0.], 2025-09-09T15:16:56.1893438Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:16:56.1893754Z converted model pt2e: GraphModule( 2025-09-09T15:16:56.1894014Z (conv): Module() 2025-09-09T15:16:56.1894224Z (bn): Module() 2025-09-09T15:16:56.1894415Z ) 2025-09-09T15:16:56.1894521Z 2025-09-09T15:16:56.1894704Z 2025-09-09T15:16:56.1894709Z 2025-09-09T15:16:56.1894796Z def forward(self, x): 2025-09-09T15:16:56.1895081Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:56.1895806Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:56.1897043Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:56.1897916Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:16:56.1898712Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.001507229870185256, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:16:56.1899505Z conv_weight_bias = self.conv.weight_bias 2025-09-09T15:16:56.1900338Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_weight_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_weight_bias = None 2025-09-09T15:16:56.1901233Z relu = torch.ops.aten.relu.default(conv2d_2); conv2d_2 = None 2025-09-09T15:16:56.1901998Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.007779817562550306, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:16:56.1903267Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.007779817562550306, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:16:56.1904273Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:16:56.1904686Z 2025-09-09T15:16:56.1904970Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:56.1905345Z onverted model fx: GraphModule( 2025-09-09T15:16:56.1905609Z (conv): ConvReLU2d( 2025-09-09T15:16:56.1905944Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:16:56.1906313Z (1): ReLU() 2025-09-09T15:16:56.1906500Z ) 2025-09-09T15:16:56.1906677Z ) 2025-09-09T15:16:56.1906773Z 2025-09-09T15:16:56.1906777Z 2025-09-09T15:16:56.1906781Z 2025-09-09T15:16:56.1906865Z def forward(self, x): 2025-09-09T15:16:56.1907480Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:56.1908706Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:56.1909706Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:16:56.1910567Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007779817562550306, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:16:56.1911838Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007779817562550306, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:56.1912710Z return dequantize_per_tensor_default_1 2025-09-09T15:16:56.1912991Z 2025-09-09T15:16:56.1913263Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:56.1913639Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:56.1913964Z [0., 0., 0.], 2025-09-09T15:16:56.1914174Z [0., 0., 0.]], 2025-09-09T15:16:56.1914312Z 2025-09-09T15:16:56.1914393Z [[0., 0., 0.], 2025-09-09T15:16:56.1914594Z [0., 0., 0.], 2025-09-09T15:16:56.1914799Z [0., 0., 0.]], 2025-09-09T15:16:56.1914934Z 2025-09-09T15:16:56.1915095Z [[0., 0., 0.], 2025-09-09T15:16:56.1915304Z [0., 0., 0.], 2025-09-09T15:16:56.1915510Z [0., 0., 0.]]]]) 2025-09-09T15:16:56.1915947Z PASSED 2025-09-09T15:16:56.1916635Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_no_bias model pt2e: GraphModule( 2025-09-09T15:16:56.1917233Z (conv): Module() 2025-09-09T15:16:56.1917538Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1918517Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:16:56.1919810Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T15:16:56.1920445Z ) 2025-09-09T15:16:56.1920734Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1921667Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:56.1922739Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:56.1923240Z ) 2025-09-09T15:16:56.1923507Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1924435Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:56.1925484Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5882599353790283) 2025-09-09T15:16:56.1925946Z ) 2025-09-09T15:16:56.1926114Z ) 2025-09-09T15:16:56.1926214Z 2025-09-09T15:16:56.1926219Z 2025-09-09T15:16:56.1926223Z 2025-09-09T15:16:56.1926307Z def forward(self, x): 2025-09-09T15:16:56.1926590Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:56.1926922Z conv_weight = self.conv.weight 2025-09-09T15:16:56.1927381Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:16:56.1927964Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:56.1928759Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:16:56.1929515Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T15:16:56.1929992Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:16:56.1930536Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:16:56.1930929Z 2025-09-09T15:16:56.1931205Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:56.1931569Z model fx: GraphModule( 2025-09-09T15:16:56.1931886Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:56.1932820Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:56.1933893Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:56.1934533Z ) 2025-09-09T15:16:56.1934719Z (conv): ConvReLU2d( 2025-09-09T15:16:56.1934969Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:16:56.1935331Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7215092Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:16:57.7216839Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1860, -0.1897, -0.1787]), max_val=tensor([0.1824, 0.1870, 0.1478])) 2025-09-09T15:16:57.7217657Z ) 2025-09-09T15:16:57.7217871Z ) 2025-09-09T15:16:57.7218201Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7219400Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:57.7220762Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5882599353790283) 2025-09-09T15:16:57.7221343Z ) 2025-09-09T15:16:57.7221550Z ) 2025-09-09T15:16:57.7221674Z 2025-09-09T15:16:57.7221680Z 2025-09-09T15:16:57.7221685Z 2025-09-09T15:16:57.7221786Z def forward(self, x): 2025-09-09T15:16:57.7222214Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:57.7222870Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:16:57.7223541Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:16:57.7224073Z return activation_post_process_1 2025-09-09T15:16:57.7224384Z 2025-09-09T15:16:57.7224723Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:57.7225178Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:57.7225465Z [0., 0., 0.], 2025-09-09T15:16:57.7225716Z [0., 0., 0.]], 2025-09-09T15:16:57.7225885Z 2025-09-09T15:16:57.7225976Z [[0., 0., 0.], 2025-09-09T15:16:57.7226228Z [0., 0., 0.], 2025-09-09T15:16:57.7226469Z [0., 0., 0.]], 2025-09-09T15:16:57.7226649Z 2025-09-09T15:16:57.7226746Z [[0., 0., 0.], 2025-09-09T15:16:57.7226988Z [0., 0., 0.], 2025-09-09T15:16:57.7227270Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:16:57.7227636Z converted model pt2e: GraphModule( 2025-09-09T15:16:57.7227955Z (conv): Module() 2025-09-09T15:16:57.7228184Z ) 2025-09-09T15:16:57.7228307Z 2025-09-09T15:16:57.7228312Z 2025-09-09T15:16:57.7228317Z 2025-09-09T15:16:57.7228417Z def forward(self, x): 2025-09-09T15:16:57.7228758Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:57.7229150Z _scale_0 = self._scale_0 2025-09-09T15:16:57.7229464Z _zero_point_0 = self._zero_point_0 2025-09-09T15:16:57.7229859Z quantize_per_channel_default = self._frozen_param0 2025-09-09T15:16:57.7231108Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T15:16:57.7232765Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:57.7234301Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:57.7235958Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T15:16:57.7237266Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T15:16:57.7238223Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.006228470243513584, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:16:57.7240004Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.006228470243513584, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:57.7241261Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:16:57.7241769Z 2025-09-09T15:16:57.7242104Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:57.7242551Z onverted model fx: GraphModule( 2025-09-09T15:16:57.7242856Z (conv): ConvReLU2d( 2025-09-09T15:16:57.7243298Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T15:16:57.7243815Z (1): ReLU() 2025-09-09T15:16:57.7244060Z ) 2025-09-09T15:16:57.7244273Z ) 2025-09-09T15:16:57.7244394Z 2025-09-09T15:16:57.7244399Z 2025-09-09T15:16:57.7244404Z 2025-09-09T15:16:57.7244513Z def forward(self, x): 2025-09-09T15:16:57.7245288Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:57.7246513Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:57.7247503Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:16:57.7248352Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.006228470243513584, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:16:57.7249622Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.006228470243513584, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:57.7250490Z return dequantize_per_tensor_default_1 2025-09-09T15:16:57.7250763Z 2025-09-09T15:16:57.7251038Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:57.7251402Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:57.7251632Z [0., 0., 0.], 2025-09-09T15:16:57.7251843Z [0., 0., 0.]], 2025-09-09T15:16:57.7251978Z 2025-09-09T15:16:57.7252058Z [[0., 0., 0.], 2025-09-09T15:16:57.7252258Z [0., 0., 0.], 2025-09-09T15:16:57.7252467Z [0., 0., 0.]], 2025-09-09T15:16:57.7252602Z 2025-09-09T15:16:57.7252675Z [[0., 0., 0.], 2025-09-09T15:16:57.7252886Z [0., 0., 0.], 2025-09-09T15:16:57.7253089Z [0., 0., 0.]]]]) 2025-09-09T15:16:57.7253320Z model pt2e: GraphModule( 2025-09-09T15:16:57.7253550Z (conv): Module() 2025-09-09T15:16:57.7253851Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7254791Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:16:57.7255889Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965116143226624, max_val=0.18703685700893402) 2025-09-09T15:16:57.7256403Z ) 2025-09-09T15:16:57.7256675Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7257600Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:57.7258678Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:57.7259275Z ) 2025-09-09T15:16:57.7259549Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7262081Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:57.7263154Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5892566442489624) 2025-09-09T15:16:57.7263626Z ) 2025-09-09T15:16:57.7264082Z ) 2025-09-09T15:16:57.7264183Z 2025-09-09T15:16:57.7264187Z 2025-09-09T15:16:57.7264191Z 2025-09-09T15:16:57.7264282Z def forward(self, x): 2025-09-09T15:16:57.7264559Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:57.7264897Z conv_weight = self.conv.weight 2025-09-09T15:16:57.7265345Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:16:57.7265932Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:57.7266729Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:16:57.7267478Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T15:16:57.7267964Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:16:57.7268501Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:16:57.7268889Z 2025-09-09T15:16:57.7269168Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:57.7269526Z model fx: GraphModule( 2025-09-09T15:16:57.7269851Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7270772Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:57.7271864Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:57.7272359Z ) 2025-09-09T15:16:57.7272546Z (conv): ConvReLU2d( 2025-09-09T15:16:57.7272807Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:16:57.7273161Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7274063Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:16:57.7275162Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.18965116143226624, max_val=0.18703685700893402) 2025-09-09T15:16:57.7275677Z ) 2025-09-09T15:16:57.7275847Z ) 2025-09-09T15:16:57.7276123Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:57.7277537Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0062]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:57.7279030Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=1.5892566442489624) 2025-09-09T15:16:57.7279500Z ) 2025-09-09T15:16:57.7279665Z ) 2025-09-09T15:16:57.7279765Z 2025-09-09T15:16:57.7279777Z 2025-09-09T15:16:57.7279781Z 2025-09-09T15:16:57.7279866Z def forward(self, x): 2025-09-09T15:16:57.7280207Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:57.7280736Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:16:57.7281280Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:16:59.4574879Z return activation_post_process_1 2025-09-09T15:16:59.4575294Z 2025-09-09T15:16:59.4575639Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:59.4576105Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:59.4576393Z [0., 0., 0.], 2025-09-09T15:16:59.4577102Z [0., 0., 0.]], 2025-09-09T15:16:59.4577278Z 2025-09-09T15:16:59.4577378Z [[0., 0., 0.], 2025-09-09T15:16:59.4577624Z [0., 0., 0.], 2025-09-09T15:16:59.4577874Z [0., 0., 0.]], 2025-09-09T15:16:59.4578040Z 2025-09-09T15:16:59.4578130Z [[0., 0., 0.], 2025-09-09T15:16:59.4578380Z [0., 0., 0.], 2025-09-09T15:16:59.4578665Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:16:59.4579038Z converted model pt2e: GraphModule( 2025-09-09T15:16:59.4579349Z (conv): Module() 2025-09-09T15:16:59.4579588Z ) 2025-09-09T15:16:59.4579706Z 2025-09-09T15:16:59.4579711Z 2025-09-09T15:16:59.4579730Z 2025-09-09T15:16:59.4579839Z def forward(self, x): 2025-09-09T15:16:59.4580170Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:59.4580630Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T15:16:59.4581777Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0014933162601664662, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:59.4583348Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:59.4584930Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:59.4586574Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:16:59.4587588Z relu = torch.ops.aten.relu.default(conv2d); conv2d = None 2025-09-09T15:16:59.4588531Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.006232379004359245, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:16:59.4590111Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.006232379004359245, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:16:59.4591359Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:16:59.4591853Z 2025-09-09T15:16:59.4592192Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:59.4592644Z onverted model fx: GraphModule( 2025-09-09T15:16:59.4592962Z (conv): ConvReLU2d( 2025-09-09T15:16:59.4593411Z (0): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T15:16:59.4593908Z (1): ReLU() 2025-09-09T15:16:59.4594141Z ) 2025-09-09T15:16:59.4594338Z ) 2025-09-09T15:16:59.4594461Z 2025-09-09T15:16:59.4594467Z 2025-09-09T15:16:59.4594472Z 2025-09-09T15:16:59.4594573Z def forward(self, x): 2025-09-09T15:16:59.4595338Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:16:59.4597011Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:16:59.4598287Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:16:59.4599358Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.006232379004359245, -128, -128, 127, torch.int8); conv = None 2025-09-09T15:16:59.4601152Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.006232379004359245, -128, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:16:59.4602241Z return dequantize_per_tensor_default_1 2025-09-09T15:16:59.4602703Z 2025-09-09T15:16:59.4603040Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:59.4603477Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:59.4603763Z [0., 0., 0.], 2025-09-09T15:16:59.4604014Z [0., 0., 0.]], 2025-09-09T15:16:59.4604192Z 2025-09-09T15:16:59.4604283Z [[0., 0., 0.], 2025-09-09T15:16:59.4604524Z [0., 0., 0.], 2025-09-09T15:16:59.4604775Z [0., 0., 0.]], 2025-09-09T15:16:59.4604940Z 2025-09-09T15:16:59.4605043Z [[0., 0., 0.], 2025-09-09T15:16:59.4605286Z [0., 0., 0.], 2025-09-09T15:16:59.4605541Z [0., 0., 0.]]]]) 2025-09-09T15:16:59.4605820Z model pt2e: GraphModule( 2025-09-09T15:16:59.4606096Z (conv): Module() 2025-09-09T15:16:59.4606452Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:59.4607662Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:16:59.4609260Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T15:16:59.4610046Z ) 2025-09-09T15:16:59.4610376Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:59.4611530Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:59.4612903Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:59.4613535Z ) 2025-09-09T15:16:59.4613857Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:59.4615023Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0080]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:59.4616393Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9872972965240479, max_val=1.0470484495162964) 2025-09-09T15:16:59.4617065Z ) 2025-09-09T15:16:59.4617268Z ) 2025-09-09T15:16:59.4617382Z 2025-09-09T15:16:59.4617387Z 2025-09-09T15:16:59.4617392Z 2025-09-09T15:16:59.4617494Z def forward(self, x): 2025-09-09T15:16:59.4617834Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:59.4618246Z conv_weight = self.conv.weight 2025-09-09T15:16:59.4618793Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:16:59.4619506Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:59.4620489Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:16:59.4621505Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T15:16:59.4622189Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:16:59.4622651Z 2025-09-09T15:16:59.4622984Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:59.4623415Z model fx: GraphModule( 2025-09-09T15:16:59.4623803Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:59.4624940Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:59.4626388Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:16:59.4627087Z ) 2025-09-09T15:16:59.4627294Z (conv): Conv2d( 2025-09-09T15:16:59.4627592Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:16:59.4628024Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:59.4629212Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0014, 0.0015]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:16:59.4630893Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1913, -0.1469, -0.1921]), max_val=tensor([0.1740, 0.1746, 0.1810])) 2025-09-09T15:16:59.4631665Z ) 2025-09-09T15:16:59.4631841Z ) 2025-09-09T15:16:59.4632113Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:16:59.4633053Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0080]), zero_point=tensor([-4], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:16:59.4634143Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9872972965240479, max_val=1.0470484495162964) 2025-09-09T15:16:59.4634655Z ) 2025-09-09T15:16:59.4634826Z ) 2025-09-09T15:16:59.4634922Z 2025-09-09T15:16:59.4634926Z 2025-09-09T15:16:59.4634930Z 2025-09-09T15:16:59.4635015Z def forward(self, x): 2025-09-09T15:16:59.4635370Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:16:59.4635896Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:16:59.4636504Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:16:59.4636934Z return activation_post_process_1 2025-09-09T15:16:59.4637188Z 2025-09-09T15:16:59.4637467Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:16:59.4637836Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:16:59.4638076Z [0., 0., 0.], 2025-09-09T15:16:59.4638283Z [0., 0., 0.]], 2025-09-09T15:16:59.4638427Z 2025-09-09T15:16:59.4638503Z [[0., 0., 0.], 2025-09-09T15:16:59.4638705Z [0., 0., 0.], 2025-09-09T15:16:59.4638915Z [0., 0., 0.]], 2025-09-09T15:16:59.4639055Z 2025-09-09T15:16:59.4639139Z [[0., 0., 0.], 2025-09-09T15:16:59.4639338Z [0., 0., 0.], 2025-09-09T15:16:59.4639576Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:16:59.4639876Z converted model pt2e: GraphModule( 2025-09-09T15:16:59.4640139Z (conv): Module() 2025-09-09T15:16:59.4640335Z ) 2025-09-09T15:16:59.4640440Z 2025-09-09T15:16:59.4640444Z 2025-09-09T15:16:59.4640448Z 2025-09-09T15:16:59.4640529Z def forward(self, x): 2025-09-09T15:16:59.4640805Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:16:59.4641133Z _scale_0 = self._scale_0 2025-09-09T15:16:59.4641392Z _zero_point_0 = self._zero_point_0 2025-09-09T15:16:59.4641711Z quantize_per_channel_default = self._frozen_param0 2025-09-09T15:16:59.4642708Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T15:16:59.4644028Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:17:00.9750254Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:17:00.9752392Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel_default); dequantize_per_tensor_default = dequantize_per_channel_default = None 2025-09-09T15:17:00.9754006Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.007977825589478016, -4, -128, 127, torch.int8); conv2d = None 2025-09-09T15:17:00.9755608Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007977825589478016, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:17:00.9756960Z return pytree.tree_unflatten((dequantize_per_tensor_default_1,), self._out_spec) 2025-09-09T15:17:00.9757465Z 2025-09-09T15:17:00.9757814Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:17:00.9758281Z onverted model fx: GraphModule( 2025-09-09T15:17:00.9758791Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T15:17:00.9759303Z ) 2025-09-09T15:17:00.9759427Z 2025-09-09T15:17:00.9759432Z 2025-09-09T15:17:00.9759437Z 2025-09-09T15:17:00.9759547Z def forward(self, x): 2025-09-09T15:17:00.9760317Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:17:00.9761873Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:17:00.9763148Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:17:00.9764574Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007977825589478016, -4, -128, 127, torch.int8); conv = None 2025-09-09T15:17:00.9766184Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007977825589478016, -4, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:17:00.9767299Z return dequantize_per_tensor_default_1 2025-09-09T15:17:00.9767624Z 2025-09-09T15:17:00.9767961Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:17:00.9768406Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:17:00.9768694Z [0., 0., 0.], 2025-09-09T15:17:00.9768949Z [0., 0., 0.]], 2025-09-09T15:17:00.9769118Z 2025-09-09T15:17:00.9769209Z [[0., 0., 0.], 2025-09-09T15:17:00.9769461Z [0., 0., 0.], 2025-09-09T15:17:00.9769705Z [0., 0., 0.]], 2025-09-09T15:17:00.9769870Z 2025-09-09T15:17:00.9769969Z [[0., 0., 0.], 2025-09-09T15:17:00.9770218Z [0., 0., 0.], 2025-09-09T15:17:00.9770471Z [0., 0., 0.]]]]) 2025-09-09T15:17:00.9770749Z model pt2e: GraphModule( 2025-09-09T15:17:00.9771028Z (conv): Module() 2025-09-09T15:17:00.9771388Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:17:00.9772578Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:17:00.9773998Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212539494037628, max_val=0.18097467720508575) 2025-09-09T15:17:00.9774639Z ) 2025-09-09T15:17:00.9774972Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:17:00.9776128Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:17:00.9777649Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:17:00.9778282Z ) 2025-09-09T15:17:00.9778606Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:17:00.9779872Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0079]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:17:00.9781300Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9800506234169006, max_val=1.0470484495162964) 2025-09-09T15:17:00.9781938Z ) 2025-09-09T15:17:00.9782142Z ) 2025-09-09T15:17:00.9782257Z 2025-09-09T15:17:00.9782262Z 2025-09-09T15:17:00.9782267Z 2025-09-09T15:17:00.9782368Z def forward(self, x): 2025-09-09T15:17:00.9782708Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:17:00.9783120Z conv_weight = self.conv.weight 2025-09-09T15:17:00.9783680Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:17:00.9784402Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:17:00.9785406Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1); activation_post_process_0 = activation_post_process_1 = None 2025-09-09T15:17:00.9786444Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T15:17:00.9787123Z return pytree.tree_unflatten((activation_post_process_2,), self._out_spec) 2025-09-09T15:17:00.9787602Z 2025-09-09T15:17:00.9787929Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:17:00.9788369Z model fx: GraphModule( 2025-09-09T15:17:00.9788757Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:17:00.9789904Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:17:00.9791256Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:17:00.9800495Z ) 2025-09-09T15:17:00.9800744Z (conv): Conv2d( 2025-09-09T15:17:00.9801056Z 3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False 2025-09-09T15:17:00.9801496Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:17:00.9802656Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:17:00.9804106Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19212539494037628, max_val=0.18097467720508575) 2025-09-09T15:17:00.9804772Z ) 2025-09-09T15:17:00.9804988Z ) 2025-09-09T15:17:00.9805319Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:17:00.9806518Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0079]), zero_point=tensor([-5], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:17:00.9807649Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.9800506234169006, max_val=1.0470484495162964) 2025-09-09T15:17:00.9808185Z ) 2025-09-09T15:17:00.9808356Z ) 2025-09-09T15:17:00.9808452Z 2025-09-09T15:17:00.9808456Z 2025-09-09T15:17:00.9808460Z 2025-09-09T15:17:00.9808545Z def forward(self, x): 2025-09-09T15:17:00.9808900Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:17:00.9809434Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:17:00.9810116Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:17:00.9810542Z return activation_post_process_1 2025-09-09T15:17:00.9810792Z 2025-09-09T15:17:00.9811068Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:17:00.9811436Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:17:00.9811755Z [0., 0., 0.], 2025-09-09T15:17:00.9811971Z [0., 0., 0.]], 2025-09-09T15:17:00.9812109Z 2025-09-09T15:17:00.9812183Z [[0., 0., 0.], 2025-09-09T15:17:00.9812389Z [0., 0., 0.], 2025-09-09T15:17:00.9812588Z [0., 0., 0.]], 2025-09-09T15:17:00.9812728Z 2025-09-09T15:17:00.9812801Z [[0., 0., 0.], 2025-09-09T15:17:00.9812995Z [0., 0., 0.], 2025-09-09T15:17:00.9813232Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:17:00.9813534Z converted model pt2e: GraphModule( 2025-09-09T15:17:00.9813801Z (conv): Module() 2025-09-09T15:17:00.9813991Z ) 2025-09-09T15:17:00.9814090Z 2025-09-09T15:17:00.9814094Z 2025-09-09T15:17:00.9814098Z 2025-09-09T15:17:00.9814179Z def forward(self, x): 2025-09-09T15:17:00.9814463Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:17:00.9814827Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T15:17:00.9815738Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.0015127983642742038, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:17:00.9816977Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:17:00.9818214Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:17:00.9819542Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_1, dequantize_per_tensor_default); dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:17:00.9820721Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.007949408143758774, -5, -128, 127, torch.int8); conv2d = None 2025-09-09T15:17:00.9821992Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.007949408143758774, -5, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:17:00.9822996Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:17:00.9823399Z 2025-09-09T15:17:00.9823676Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:17:00.9824048Z onverted model fx: GraphModule( 2025-09-09T15:17:00.9824456Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1), bias=False) 2025-09-09T15:17:00.9824875Z ) 2025-09-09T15:17:00.9824970Z 2025-09-09T15:17:00.9824974Z 2025-09-09T15:17:00.9824977Z 2025-09-09T15:17:00.9825057Z def forward(self, x): 2025-09-09T15:17:00.9825666Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:18:21.2765705Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:18:21.2766747Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:18:21.2767592Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.007949408143758774, -5, -128, 127, torch.int8); conv = None 2025-09-09T15:18:21.2768864Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.007949408143758774, -5, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:18:21.2770208Z return dequantize_per_tensor_default_1 2025-09-09T15:18:21.2770485Z 2025-09-09T15:18:21.2770801Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:21.2771341Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:18:21.2771581Z [0., 0., 0.], 2025-09-09T15:18:21.2771788Z [0., 0., 0.]], 2025-09-09T15:18:21.2771935Z 2025-09-09T15:18:21.2772009Z [[0., 0., 0.], 2025-09-09T15:18:21.2772206Z [0., 0., 0.], 2025-09-09T15:18:21.2772411Z [0., 0., 0.]], 2025-09-09T15:18:21.2772548Z 2025-09-09T15:18:21.2772628Z [[0., 0., 0.], 2025-09-09T15:18:21.2772824Z [0., 0., 0.], 2025-09-09T15:18:21.2773041Z [0., 0., 0.]]]]) 2025-09-09T15:18:21.2773491Z PASSED 2025-09-09T15:18:21.2774141Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_transpose_bn PASSED 2025-09-09T15:18:21.2775160Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_transpose_bn_relu PASSED 2025-09-09T15:18:21.2776104Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_inplace_add_relu model pt2e: GraphModule( 2025-09-09T15:18:21.2776706Z (conv): Module() 2025-09-09T15:18:21.2777004Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2777964Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:18:21.2779144Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2877]), max_val=tensor([-0.2877])) 2025-09-09T15:18:21.2779697Z ) 2025-09-09T15:18:21.2779974Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2780892Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:21.2781979Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T15:18:21.2782477Z ) 2025-09-09T15:18:21.2782755Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2783675Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:21.2784751Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T15:18:21.2785260Z ) 2025-09-09T15:18:21.2785537Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2786462Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:21.2787505Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T15:18:21.2787956Z ) 2025-09-09T15:18:21.2788130Z ) 2025-09-09T15:18:21.2788222Z 2025-09-09T15:18:21.2788227Z 2025-09-09T15:18:21.2788231Z 2025-09-09T15:18:21.2788313Z def forward(self, x): 2025-09-09T15:18:21.2788593Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:18:21.2788928Z conv_weight = self.conv.weight 2025-09-09T15:18:21.2789379Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:18:21.2789842Z conv_bias = self.conv.bias 2025-09-09T15:18:21.2790357Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:18:21.2791134Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T15:18:21.2792013Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T15:18:21.2792807Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T15:18:21.2793507Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:18:21.2793974Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T15:18:21.2794517Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:18:21.2794893Z 2025-09-09T15:18:21.2795171Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:21.2795544Z model fx: GraphModule( 2025-09-09T15:18:21.2795856Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2796877Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:21.2797960Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T15:18:21.2798467Z ) 2025-09-09T15:18:21.2798642Z (conv): Conv2d( 2025-09-09T15:18:21.2798886Z 1, 1, kernel_size=(1, 1), stride=(1, 1) 2025-09-09T15:18:21.2799225Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2800134Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:18:21.2801301Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.2877]), max_val=tensor([-0.2877])) 2025-09-09T15:18:21.2801859Z ) 2025-09-09T15:18:21.2802024Z ) 2025-09-09T15:18:21.2802296Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2803217Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:21.2804306Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T15:18:21.2804810Z ) 2025-09-09T15:18:21.2804999Z (relu): ReLU(inplace=True) 2025-09-09T15:18:21.2805336Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:21.2806262Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:21.2807311Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T15:18:21.2807774Z ) 2025-09-09T15:18:21.2807944Z ) 2025-09-09T15:18:21.2808040Z 2025-09-09T15:18:21.2808045Z 2025-09-09T15:18:21.2808049Z 2025-09-09T15:18:21.2808138Z def forward(self, x): 2025-09-09T15:18:21.2808479Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:18:21.2808902Z conv = self.conv(activation_post_process_0) 2025-09-09T15:18:21.2809329Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:18:21.2810012Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T15:18:21.2810560Z relu = self.relu(add); add = None 2025-09-09T15:18:21.2811077Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:18:21.2811498Z return activation_post_process_2 2025-09-09T15:18:21.2811749Z 2025-09-09T15:18:21.2812022Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:21.2812462Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:18:21.2812703Z [0., 0., 0.], 2025-09-09T15:18:21.2812939Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:18:21.2813244Z converted model pt2e: GraphModule( 2025-09-09T15:18:21.2813500Z (conv): Module() 2025-09-09T15:18:21.2813695Z ) 2025-09-09T15:18:21.2813789Z 2025-09-09T15:18:21.2813794Z 2025-09-09T15:18:21.2813797Z 2025-09-09T15:18:21.2813890Z def forward(self, x): 2025-09-09T15:18:21.2814165Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:18:21.2814492Z _scale_0 = self._scale_0 2025-09-09T15:18:21.2814737Z _zero_point_0 = self._zero_point_0 2025-09-09T15:18:21.2815069Z quantize_per_channel_default = self._frozen_param0 2025-09-09T15:18:21.2816052Z dequantize_per_channel_default = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel_default, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel_default = _scale_0 = _zero_point_0 = None 2025-09-09T15:18:21.2816999Z conv_bias = self.conv.bias 2025-09-09T15:18:21.2817622Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T15:18:21.2818713Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8) 2025-09-09T15:18:21.2820010Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:18:21.2821400Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_3, dequantize_per_channel_default, conv_bias); dequantize_per_tensor_default_3 = dequantize_per_channel_default = conv_bias = None 2025-09-09T15:18:21.2822634Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.003615050343796611, 43, -128, 127, torch.int8); conv2d = None 2025-09-09T15:18:22.9220941Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:18:22.9222272Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_1, dequantize_per_tensor_default_4); dequantize_per_tensor_default_1 = dequantize_per_tensor_default_4 = None 2025-09-09T15:18:22.9223059Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:18:22.9223854Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.003489378374069929, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T15:18:22.9225146Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:18:22.9226154Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:18:22.9226560Z 2025-09-09T15:18:22.9226847Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:22.9227218Z onverted model fx: GraphModule( 2025-09-09T15:18:22.9227609Z (conv): QuantizedConv2d(Reference)(1, 1, kernel_size=(1, 1), stride=(1, 1)) 2025-09-09T15:18:22.9228036Z (relu): ReLU(inplace=True) 2025-09-09T15:18:22.9228269Z ) 2025-09-09T15:18:22.9228376Z 2025-09-09T15:18:22.9228380Z 2025-09-09T15:18:22.9228384Z 2025-09-09T15:18:22.9228881Z def forward(self, x): 2025-09-09T15:18:22.9229492Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T15:18:22.9230845Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:18:22.9231729Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T15:18:22.9232445Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.003615050343796611, 43, -128, 127, torch.int8); conv = None 2025-09-09T15:18:22.9233704Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:18:22.9234897Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:18:22.9235525Z relu = self.relu(add); add = None 2025-09-09T15:18:22.9236223Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.003489378374069929, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:18:22.9238234Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:18:22.9239486Z return dequantize_per_tensor_default_2 2025-09-09T15:18:22.9239759Z 2025-09-09T15:18:22.9240032Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:22.9240405Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:18:22.9240638Z [0., 0., 0.], 2025-09-09T15:18:22.9240854Z [0., 0., 0.]]]]) 2025-09-09T15:18:22.9241090Z model pt2e: GraphModule( 2025-09-09T15:18:22.9241320Z (conv): Module() 2025-09-09T15:18:22.9241627Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9242571Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:18:22.9243714Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.28767645359039307, max_val=-0.28767645359039307) 2025-09-09T15:18:22.9244232Z ) 2025-09-09T15:18:22.9244503Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9245426Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:22.9246508Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T15:18:22.9247020Z ) 2025-09-09T15:18:22.9247295Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9248216Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:22.9249300Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T15:18:22.9249805Z ) 2025-09-09T15:18:22.9250076Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9251005Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:22.9252149Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T15:18:22.9252616Z ) 2025-09-09T15:18:22.9252786Z ) 2025-09-09T15:18:22.9252888Z 2025-09-09T15:18:22.9252892Z 2025-09-09T15:18:22.9252896Z 2025-09-09T15:18:22.9252979Z def forward(self, x): 2025-09-09T15:18:22.9253339Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:18:22.9253671Z conv_weight = self.conv.weight 2025-09-09T15:18:22.9254128Z activation_post_process_1 = self.activation_post_process_1(conv_weight); conv_weight = None 2025-09-09T15:18:22.9254589Z conv_bias = self.conv.bias 2025-09-09T15:18:22.9254965Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:18:22.9255734Z conv2d = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, conv_bias); activation_post_process_1 = conv_bias = None 2025-09-09T15:18:22.9256538Z activation_post_process_2 = self.activation_post_process_2(conv2d); conv2d = None 2025-09-09T15:18:22.9257340Z add_ = torch.ops.aten.add_.Tensor(activation_post_process_2, activation_post_process_0); activation_post_process_2 = activation_post_process_0 = None 2025-09-09T15:18:22.9258040Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:18:22.9258523Z activation_post_process_3 = self.activation_post_process_3(relu_); relu_ = None 2025-09-09T15:18:22.9259065Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:18:22.9259455Z 2025-09-09T15:18:22.9259734Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:22.9260102Z model fx: GraphModule( 2025-09-09T15:18:22.9260425Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9261359Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0126]), zero_point=tensor([7], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:22.9262472Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-1.7008640766143799, max_val=1.5035617351531982) 2025-09-09T15:18:22.9262978Z ) 2025-09-09T15:18:22.9263160Z (conv): Conv2d( 2025-09-09T15:18:22.9263391Z 1, 1, kernel_size=(1, 1), stride=(1, 1) 2025-09-09T15:18:22.9264262Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9265521Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0011]), zero_point=tensor([127], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:18:22.9266956Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.28767645359039307, max_val=-0.28767645359039307) 2025-09-09T15:18:22.9267478Z ) 2025-09-09T15:18:22.9267644Z ) 2025-09-09T15:18:22.9267924Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9268872Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0036]), zero_point=tensor([43], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:22.9269981Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.6198297739028931, max_val=0.30200809240341187) 2025-09-09T15:18:22.9270494Z ) 2025-09-09T15:18:22.9270679Z (relu): ReLU(inplace=True) 2025-09-09T15:18:22.9271023Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:18:22.9271969Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0035]), zero_point=tensor([-128], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:18:22.9273026Z (activation_post_process): MovingAverageMinMaxObserver(min_val=0.0, max_val=0.8897914886474609) 2025-09-09T15:18:22.9273685Z ) 2025-09-09T15:18:22.9273845Z ) 2025-09-09T15:18:22.9273947Z 2025-09-09T15:18:22.9273950Z 2025-09-09T15:18:22.9273954Z 2025-09-09T15:18:22.9274038Z def forward(self, x): 2025-09-09T15:18:22.9274389Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:18:22.9274928Z conv = self.conv(activation_post_process_0) 2025-09-09T15:18:22.9275372Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:18:22.9276057Z add = activation_post_process_1 + activation_post_process_0; activation_post_process_1 = activation_post_process_0 = None 2025-09-09T15:18:22.9276879Z relu = self.relu(add); add = None 2025-09-09T15:18:22.9277444Z activation_post_process_2 = self.activation_post_process_2(relu); relu = None 2025-09-09T15:18:22.9278056Z return activation_post_process_2 2025-09-09T15:18:22.9278408Z 2025-09-09T15:18:22.9278750Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:18:22.9279126Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:18:22.9279359Z [0., 0., 0.], 2025-09-09T15:18:22.9279606Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:18:22.9279910Z converted model pt2e: GraphModule( 2025-09-09T15:18:22.9280170Z (conv): Module() 2025-09-09T15:18:22.9280361Z ) 2025-09-09T15:18:22.9280462Z 2025-09-09T15:18:22.9280472Z 2025-09-09T15:18:22.9280476Z 2025-09-09T15:18:22.9280563Z def forward(self, x): 2025-09-09T15:18:22.9280845Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:18:22.9281215Z quantize_per_tensor_default = self._frozen_param0 2025-09-09T15:19:06.0494908Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.002265168819576502, 0, -127, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:19:06.0496914Z conv_bias = self.conv.bias 2025-09-09T15:19:06.0497960Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T15:19:06.0499443Z dequantize_per_tensor_default_5 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01256637554615736, 7, -128, 127, torch.int8) 2025-09-09T15:19:06.0501177Z dequantize_per_tensor_default_4 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:19:06.0502964Z conv2d = torch.ops.aten.conv2d.default(dequantize_per_tensor_default_4, dequantize_per_tensor_default, conv_bias); dequantize_per_tensor_default_4 = dequantize_per_tensor_default = conv_bias = None 2025-09-09T15:19:06.0504561Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d, 0.003615050343796611, 43, -128, 127, torch.int8); conv2d = None 2025-09-09T15:19:06.0506204Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:19:06.0507882Z add_ = torch.ops.aten.add_.Tensor(dequantize_per_tensor_default_2, dequantize_per_tensor_default_5); dequantize_per_tensor_default_2 = dequantize_per_tensor_default_5 = None 2025-09-09T15:19:06.0508877Z relu_ = torch.ops.aten.relu_.default(add_); add_ = None 2025-09-09T15:19:06.0509830Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu_, 0.003489378374069929, -128, -128, 127, torch.int8); relu_ = None 2025-09-09T15:19:06.0511451Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T15:19:06.0512720Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T15:19:06.0513710Z 2025-09-09T15:19:06.0514053Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:06.0514505Z onverted model fx: GraphModule( 2025-09-09T15:19:06.0514972Z (conv): QuantizedConv2d(Reference)(1, 1, kernel_size=(1, 1), stride=(1, 1)) 2025-09-09T15:19:06.0515631Z (relu): ReLU(inplace=True) 2025-09-09T15:19:06.0515914Z ) 2025-09-09T15:19:06.0516033Z 2025-09-09T15:19:06.0516038Z 2025-09-09T15:19:06.0516043Z 2025-09-09T15:19:06.0516151Z def forward(self, x): 2025-09-09T15:19:06.0517019Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.01256637554615736, 7, -128, 127, torch.int8); x = None 2025-09-09T15:19:06.0518580Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.01256637554615736, 7, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:19:06.0519718Z conv = self.conv(dequantize_per_tensor_default) 2025-09-09T15:19:06.0520625Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.003615050343796611, 43, -128, 127, torch.int8); conv = None 2025-09-09T15:19:06.0522251Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.003615050343796611, 43, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:19:06.0523783Z add = dequantize_per_tensor_default_1 + dequantize_per_tensor_default; dequantize_per_tensor_default_1 = dequantize_per_tensor_default = None 2025-09-09T15:19:06.0524590Z relu = self.relu(add); add = None 2025-09-09T15:19:06.0525485Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(relu, 0.003489378374069929, -128, -128, 127, torch.int8); relu = None 2025-09-09T15:19:06.0527070Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.003489378374069929, -128, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:19:06.0527955Z return dequantize_per_tensor_default_2 2025-09-09T15:19:06.0528224Z 2025-09-09T15:19:06.0528500Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:06.0528879Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:19:06.0529115Z [0., 0., 0.], 2025-09-09T15:19:06.0529325Z [0., 0., 0.]]]]) 2025-09-09T15:19:06.0529753Z PASSED 2025-09-09T15:19:06.0530468Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_per_channel_weight_custom_dtype PASSED 2025-09-09T15:19:06.0531530Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_preserve_source_fn_stack PASSED 2025-09-09T15:19:06.0532484Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_update_shared_qspec model pt2e: GraphModule( 2025-09-09T15:19:06.0533101Z (conv): Module() 2025-09-09T15:19:06.0533300Z (bn): Module() 2025-09-09T15:19:06.0533597Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:06.0534529Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:06.0535697Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:19:06.0536199Z ) 2025-09-09T15:19:06.0536476Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:06.0537458Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:19:06.0538848Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1640, -0.1903, -0.1739]), max_val=tensor([0.1851, 0.1825, 0.1577])) 2025-09-09T15:19:06.0539484Z ) 2025-09-09T15:19:06.0539762Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:06.0540757Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:06.0541862Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T15:19:06.0542407Z ) 2025-09-09T15:19:06.0542675Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:06.0543596Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:06.0544674Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T15:19:06.0545185Z ) 2025-09-09T15:19:06.0545352Z ) 2025-09-09T15:19:06.0545453Z 2025-09-09T15:19:06.0545457Z 2025-09-09T15:19:06.0545466Z 2025-09-09T15:19:06.0545551Z def forward(self, x): 2025-09-09T15:19:06.0545836Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:19:06.0546166Z conv_weight = self.conv.weight 2025-09-09T15:19:06.0546436Z conv_bias = self.conv.bias 2025-09-09T15:19:06.0546683Z bn_weight = self.bn.weight 2025-09-09T15:19:06.0546928Z bn_bias = self.bn.bias 2025-09-09T15:19:06.0547178Z bn_running_mean = self.bn.running_mean 2025-09-09T15:19:06.0547476Z bn_running_var = self.bn.running_var 2025-09-09T15:19:06.0547805Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:19:06.0548242Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:19:06.0548824Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:19:06.0549341Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:19:06.0549734Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:19:06.0550137Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:19:06.0550578Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:19:06.0551081Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:19:06.0551634Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:19:06.0552243Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:19:06.0553199Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:19:06.0554093Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:19:06.0563240Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:19:06.0564096Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:19:06.0564663Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:19:06.0565656Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:19:06.0566569Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:19:06.0567304Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T15:19:06.0568255Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T15:19:24.2732319Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:19:24.2733174Z 2025-09-09T15:19:24.2733930Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:24.2734390Z model fx: GraphModule( 2025-09-09T15:19:24.2734778Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2735964Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:24.2737364Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:19:24.2738010Z ) 2025-09-09T15:19:24.2738229Z (conv): ConvBn2d( 2025-09-09T15:19:24.2738496Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:19:24.2739001Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:19:24.2739567Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2740742Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015, 0.0015, 0.0014]), zero_point=tensor([0, 0, 0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_channel_symmetric, reduce_range=False 2025-09-09T15:19:24.2742333Z (activation_post_process): MovingAveragePerChannelMinMaxObserver(min_val=tensor([-0.1640, -0.1903, -0.1739]), max_val=tensor([0.1851, 0.1825, 0.1577])) 2025-09-09T15:19:24.2743133Z ) 2025-09-09T15:19:24.2743349Z ) 2025-09-09T15:19:24.2743672Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2744818Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:24.2746175Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T15:19:24.2746810Z ) 2025-09-09T15:19:24.2747074Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:19:24.2747543Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2748684Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:24.2750032Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1757256984710693, max_val=1.9743094444274902) 2025-09-09T15:19:24.2750653Z ) 2025-09-09T15:19:24.2750866Z ) 2025-09-09T15:19:24.2750982Z 2025-09-09T15:19:24.2750988Z 2025-09-09T15:19:24.2750993Z 2025-09-09T15:19:24.2751095Z def forward(self, x): 2025-09-09T15:19:24.2751524Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:19:24.2752176Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:19:24.2752836Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:19:24.2753548Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:19:24.2754331Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T15:19:24.2754890Z return activation_post_process_2 2025-09-09T15:19:24.2755196Z 2025-09-09T15:19:24.2755536Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:24.2755989Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:19:24.2756390Z [0., 0., 0.], 2025-09-09T15:19:24.2756842Z [0., 0., 0.]], 2025-09-09T15:19:24.2757010Z 2025-09-09T15:19:24.2757103Z [[0., 0., 0.], 2025-09-09T15:19:24.2757359Z [0., 0., 0.], 2025-09-09T15:19:24.2757604Z [0., 0., 0.]], 2025-09-09T15:19:24.2757775Z 2025-09-09T15:19:24.2757866Z [[0., 0., 0.], 2025-09-09T15:19:24.2758192Z [0., 0., 0.], 2025-09-09T15:19:24.2758485Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:19:24.2758860Z converted model pt2e: GraphModule( 2025-09-09T15:19:24.2759172Z (conv): Module() 2025-09-09T15:19:24.2759418Z (bn): Module() 2025-09-09T15:19:24.2759643Z ) 2025-09-09T15:19:24.2759760Z 2025-09-09T15:19:24.2759774Z 2025-09-09T15:19:24.2759779Z 2025-09-09T15:19:24.2759880Z def forward(self, x): 2025-09-09T15:19:24.2760210Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:19:24.2760621Z conv_bias = self.conv.bias 2025-09-09T15:19:24.2761432Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:19:24.2763006Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:19:24.2764472Z _scale_0 = self._scale_0 2025-09-09T15:19:24.2764782Z _zero_point_0 = self._zero_point_0 2025-09-09T15:19:24.2765149Z quantize_per_channel = self._frozen_param0 2025-09-09T15:19:24.2766251Z dequantize_per_channel = torch.ops.quantized_decomposed.dequantize_per_channel.default(quantize_per_channel, _scale_0, _zero_point_0, 0, -127, 127, torch.int8); quantize_per_channel = _scale_0 = _zero_point_0 = None 2025-09-09T15:19:24.2767950Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_channel, conv_bias); dequantize_per_tensor_default = dequantize_per_channel = conv_bias = None 2025-09-09T15:19:24.2769542Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.016274645924568176, 6, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:19:24.2770873Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:19:24.2772032Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_1, -1.0, 1.0); dequantize_per_tensor_default_1 = None 2025-09-09T15:19:24.2773044Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.016274645924568176, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:19:24.2774323Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:19:24.2775330Z return pytree.tree_unflatten((dequantize_per_tensor_default_2,), self._out_spec) 2025-09-09T15:19:24.2775750Z 2025-09-09T15:19:24.2776033Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:24.2776415Z onverted model fx: GraphModule( 2025-09-09T15:19:24.2776798Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:19:24.2777229Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:19:24.2777520Z ) 2025-09-09T15:19:24.2777626Z 2025-09-09T15:19:24.2777630Z 2025-09-09T15:19:24.2777634Z 2025-09-09T15:19:24.2777723Z def forward(self, x): 2025-09-09T15:19:24.2778337Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:19:24.2779557Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:19:24.2780717Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:19:24.2781670Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.016274645924568176, 6, -128, 127, torch.int8); conv = None 2025-09-09T15:19:24.2782914Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:19:24.2784022Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:19:24.2784929Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.016274645924568176, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:19:24.2786202Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.016274645924568176, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:19:24.2787074Z return dequantize_per_tensor_default_2 2025-09-09T15:19:24.2787344Z 2025-09-09T15:19:24.2787633Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:24.2787999Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:19:24.2788240Z [0., 0., 0.], 2025-09-09T15:19:24.2788447Z [0., 0., 0.]], 2025-09-09T15:19:24.2788591Z 2025-09-09T15:19:24.2788666Z [[0., 0., 0.], 2025-09-09T15:19:24.2788875Z [0., 0., 0.], 2025-09-09T15:19:24.2789075Z [0., 0., 0.]], 2025-09-09T15:19:24.2789215Z 2025-09-09T15:19:24.2789295Z [[0., 0., 0.], 2025-09-09T15:19:24.2789494Z [0., 0., 0.], 2025-09-09T15:19:24.2789706Z [0., 0., 0.]]]]) 2025-09-09T15:19:24.2789934Z model pt2e: GraphModule( 2025-09-09T15:19:24.2790170Z (conv): Module() 2025-09-09T15:19:24.2790368Z (bn): Module() 2025-09-09T15:19:24.2790674Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2791610Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:24.2792692Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:19:24.2793194Z ) 2025-09-09T15:19:24.2793465Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2794402Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.int8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:19:24.2795504Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19029980897903442, max_val=0.18509264290332794) 2025-09-09T15:19:24.2796016Z ) 2025-09-09T15:19:24.2796394Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:24.2797344Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:43.5202737Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T15:19:43.5203461Z ) 2025-09-09T15:19:43.5203795Z (activation_post_process_3): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:43.5204969Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.int8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:43.5206771Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T15:19:43.5207388Z ) 2025-09-09T15:19:43.5207599Z ) 2025-09-09T15:19:43.5207713Z 2025-09-09T15:19:43.5207719Z 2025-09-09T15:19:43.5207723Z 2025-09-09T15:19:43.5207829Z def forward(self, x): 2025-09-09T15:19:43.5208344Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:19:43.5208759Z conv_weight = self.conv.weight 2025-09-09T15:19:43.5209085Z conv_bias = self.conv.bias 2025-09-09T15:19:43.5209392Z bn_weight = self.bn.weight 2025-09-09T15:19:43.5209685Z bn_bias = self.bn.bias 2025-09-09T15:19:43.5209993Z bn_running_mean = self.bn.running_mean 2025-09-09T15:19:43.5210350Z bn_running_var = self.bn.running_var 2025-09-09T15:19:43.5210747Z bn_num_batches_tracked = self.bn.num_batches_tracked 2025-09-09T15:19:43.5211292Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:19:43.5212014Z add_ = torch.ops.aten.add_.Tensor(bn_num_batches_tracked, 1); bn_num_batches_tracked = add_ = None 2025-09-09T15:19:43.5212648Z add = torch.ops.aten.add.Tensor(bn_running_var, 1e-05) 2025-09-09T15:19:43.5213108Z sqrt = torch.ops.aten.sqrt.default(add); add = None 2025-09-09T15:19:43.5213609Z div = torch.ops.aten.div.Tensor(bn_weight, sqrt); sqrt = None 2025-09-09T15:19:43.5214137Z reshape = torch.ops.aten.reshape.default(div, [-1, 1, 1, 1]) 2025-09-09T15:19:43.5214749Z mul = torch.ops.aten.mul.Tensor(conv_weight, reshape); conv_weight = reshape = None 2025-09-09T15:19:43.5215426Z activation_post_process_1 = self.activation_post_process_1(mul); mul = None 2025-09-09T15:19:43.5216185Z zeros_like = torch.ops.aten.zeros_like.default(conv_bias, dtype = torch.float32, pin_memory = False) 2025-09-09T15:19:43.5217402Z conv2d_1 = torch.ops.aten.conv2d.default(activation_post_process_0, activation_post_process_1, zeros_like); activation_post_process_0 = activation_post_process_1 = zeros_like = None 2025-09-09T15:19:43.5218488Z reshape_1 = torch.ops.aten.reshape.default(div, [1, -1, 1, 1]); div = None 2025-09-09T15:19:43.5219140Z div_1 = torch.ops.aten.div.Tensor(conv2d_1, reshape_1); conv2d_1 = reshape_1 = None 2025-09-09T15:19:43.5219851Z reshape_2 = torch.ops.aten.reshape.default(conv_bias, [1, -1, 1, 1]); conv_bias = None 2025-09-09T15:19:43.5220529Z add_1 = torch.ops.aten.add.Tensor(div_1, reshape_2); div_1 = reshape_2 = None 2025-09-09T15:19:43.5221604Z batch_norm_1 = torch.ops.aten.batch_norm.default(add_1, bn_weight, bn_bias, bn_running_mean, bn_running_var, True, 0.1, 1e-05, True); add_1 = bn_weight = bn_bias = bn_running_mean = bn_running_var = None 2025-09-09T15:19:43.5222753Z activation_post_process_2 = self.activation_post_process_2(batch_norm_1); batch_norm_1 = None 2025-09-09T15:19:43.5223665Z hardtanh = torch.ops.aten.hardtanh.default(activation_post_process_2, -1.0, 1.0); activation_post_process_2 = None 2025-09-09T15:19:43.5224533Z activation_post_process_3 = self.activation_post_process_3(hardtanh); hardtanh = None 2025-09-09T15:19:43.5225219Z return pytree.tree_unflatten((activation_post_process_3,), self._out_spec) 2025-09-09T15:19:43.5225688Z 2025-09-09T15:19:43.5226023Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:43.5226471Z model fx: GraphModule( 2025-09-09T15:19:43.5226845Z (activation_post_process_0): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:43.5227991Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0183]), zero_point=tensor([10], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:43.5229339Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.526270866394043, max_val=2.143237352371216) 2025-09-09T15:19:43.5229954Z ) 2025-09-09T15:19:43.5230168Z (conv): ConvBn2d( 2025-09-09T15:19:43.5230432Z 3, 3, kernel_size=(3, 3), stride=(1, 1) 2025-09-09T15:19:43.5231034Z (bn): BatchNorm2d(3, eps=1e-05, momentum=0.1, affine=True, track_running_stats=True) 2025-09-09T15:19:43.5231595Z (weight_fake_quant): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:43.5232816Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0015]), zero_point=tensor([0], dtype=torch.int32), dtype=torch.qint8, quant_min=-127, quant_max=127, qscheme=torch.per_tensor_symmetric, reduce_range=False 2025-09-09T15:19:43.5234221Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-0.19029980897903442, max_val=0.18509264290332794) 2025-09-09T15:19:43.5234868Z ) 2025-09-09T15:19:43.5235079Z ) 2025-09-09T15:19:43.5235401Z (activation_post_process_1): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:43.5236662Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:43.5238048Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T15:19:43.5238676Z ) 2025-09-09T15:19:43.5238940Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:19:43.5239418Z (activation_post_process_2): FusedMovingAvgObsFakeQuantize( 2025-09-09T15:19:43.5240587Z fake_quant_enabled=tensor([1]), observer_enabled=tensor([1]), scale=tensor([0.0163]), zero_point=tensor([6], dtype=torch.int32), dtype=torch.qint8, quant_min=-128, quant_max=127, qscheme=torch.per_tensor_affine, reduce_range=False 2025-09-09T15:19:43.5241948Z (activation_post_process): MovingAverageMinMaxObserver(min_val=-2.1751670837402344, max_val=1.979515790939331) 2025-09-09T15:19:43.5242594Z ) 2025-09-09T15:19:43.5242791Z ) 2025-09-09T15:19:43.5242913Z 2025-09-09T15:19:43.5242918Z 2025-09-09T15:19:43.5242923Z 2025-09-09T15:19:43.5243024Z def forward(self, x): 2025-09-09T15:19:43.5243454Z activation_post_process_0 = self.activation_post_process_0(x); x = None 2025-09-09T15:19:43.5244092Z conv = self.conv(activation_post_process_0); activation_post_process_0 = None 2025-09-09T15:19:43.5244762Z activation_post_process_1 = self.activation_post_process_1(conv); conv = None 2025-09-09T15:19:43.5245465Z hardtanh = self.hardtanh(activation_post_process_1); activation_post_process_1 = None 2025-09-09T15:19:43.5246232Z activation_post_process_2 = self.activation_post_process_2(hardtanh); hardtanh = None 2025-09-09T15:19:43.5246807Z return activation_post_process_2 2025-09-09T15:19:43.5247112Z 2025-09-09T15:19:43.5247442Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:43.5247880Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:19:43.5248166Z [0., 0., 0.], 2025-09-09T15:19:43.5248413Z [0., 0., 0.]], 2025-09-09T15:19:43.5248587Z 2025-09-09T15:19:43.5248678Z [[0., 0., 0.], 2025-09-09T15:19:43.5248929Z [0., 0., 0.], 2025-09-09T15:19:43.5249179Z [0., 0., 0.]], 2025-09-09T15:19:43.5249344Z 2025-09-09T15:19:43.5249442Z [[0., 0., 0.], 2025-09-09T15:19:43.5249688Z [0., 0., 0.], 2025-09-09T15:19:43.5249976Z [0., 0., 0.]]]], grad_fn=) 2025-09-09T15:19:43.5250340Z converted model pt2e: GraphModule( 2025-09-09T15:19:43.5250659Z (conv): Module() 2025-09-09T15:19:43.5250894Z (bn): Module() 2025-09-09T15:19:43.5251136Z ) 2025-09-09T15:19:43.5251250Z 2025-09-09T15:19:43.5251255Z 2025-09-09T15:19:43.5251260Z 2025-09-09T15:19:43.5251362Z def forward(self, x): 2025-09-09T15:19:43.5251693Z x, = fx_pytree.tree_flatten_spec(([x], {}), self._in_spec) 2025-09-09T15:19:43.5252092Z conv_bias = self.conv.bias 2025-09-09T15:19:43.5252866Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:19:43.5254415Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:19:43.5255391Z quantize_per_tensor = self._frozen_param0 2025-09-09T15:19:43.5256299Z dequantize_per_tensor = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor, 0.0014984237495809793, 0, -127, 127, torch.int8); quantize_per_tensor = None 2025-09-09T15:19:43.5257552Z conv2d_2 = torch.ops.aten.conv2d.default(dequantize_per_tensor_default, dequantize_per_tensor, conv_bias); dequantize_per_tensor_default = dequantize_per_tensor = conv_bias = None 2025-09-09T15:19:43.5258720Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv2d_2, 0.0162928756326437, 6, -128, 127, torch.int8); conv2d_2 = None 2025-09-09T15:19:43.5259981Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:19:43.5261136Z hardtanh = torch.ops.aten.hardtanh.default(dequantize_per_tensor_default_2, -1.0, 1.0); dequantize_per_tensor_default_2 = None 2025-09-09T15:19:43.5262136Z quantize_per_tensor_default_3 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.0162928756326437, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:19:43.5263400Z dequantize_per_tensor_default_3 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_3, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_3 = None 2025-09-09T15:19:43.5264648Z return pytree.tree_unflatten((dequantize_per_tensor_default_3,), self._out_spec) 2025-09-09T15:19:43.5265053Z 2025-09-09T15:19:43.5265331Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:19:43.5265700Z onverted model fx: GraphModule( 2025-09-09T15:19:43.5266087Z (conv): QuantizedConv2d(Reference)(3, 3, kernel_size=(3, 3), stride=(1, 1)) 2025-09-09T15:19:43.5266505Z (hardtanh): Hardtanh(min_val=-1.0, max_val=1.0) 2025-09-09T15:19:43.5266795Z ) 2025-09-09T15:19:43.5266893Z 2025-09-09T15:19:43.5266897Z 2025-09-09T15:19:43.5266901Z 2025-09-09T15:19:43.5266990Z def forward(self, x): 2025-09-09T15:19:43.5267601Z quantize_per_tensor_default = torch.ops.quantized_decomposed.quantize_per_tensor.default(x, 0.018311796709895134, 10, -128, 127, torch.int8); x = None 2025-09-09T15:19:43.5268827Z dequantize_per_tensor_default = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default, 0.018311796709895134, 10, -128, 127, torch.int8); quantize_per_tensor_default = None 2025-09-09T15:20:40.9766103Z conv = self.conv(dequantize_per_tensor_default); dequantize_per_tensor_default = None 2025-09-09T15:20:40.9766998Z quantize_per_tensor_default_1 = torch.ops.quantized_decomposed.quantize_per_tensor.default(conv, 0.0162928756326437, 6, -128, 127, torch.int8); conv = None 2025-09-09T15:20:40.9768286Z dequantize_per_tensor_default_1 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_1, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_1 = None 2025-09-09T15:20:40.9769358Z hardtanh = self.hardtanh(dequantize_per_tensor_default_1); dequantize_per_tensor_default_1 = None 2025-09-09T15:20:40.9770282Z quantize_per_tensor_default_2 = torch.ops.quantized_decomposed.quantize_per_tensor.default(hardtanh, 0.0162928756326437, 6, -128, 127, torch.int8); hardtanh = None 2025-09-09T15:20:40.9771580Z dequantize_per_tensor_default_2 = torch.ops.quantized_decomposed.dequantize_per_tensor.default(quantize_per_tensor_default_2, 0.0162928756326437, 6, -128, 127, torch.int8); quantize_per_tensor_default_2 = None 2025-09-09T15:20:40.9772444Z return dequantize_per_tensor_default_2 2025-09-09T15:20:40.9772724Z 2025-09-09T15:20:40.9773004Z # To see more debug info, please use `graph_module.print_readable()` 2025-09-09T15:20:40.9775187Z diff: tensor([[[[0., 0., 0.], 2025-09-09T15:20:40.9775431Z [0., 0., 0.], 2025-09-09T15:20:40.9775640Z [0., 0., 0.]], 2025-09-09T15:20:40.9775781Z 2025-09-09T15:20:40.9775862Z [[0., 0., 0.], 2025-09-09T15:20:40.9776073Z [0., 0., 0.], 2025-09-09T15:20:40.9776464Z [0., 0., 0.]], 2025-09-09T15:20:40.9776604Z 2025-09-09T15:20:40.9776680Z [[0., 0., 0.], 2025-09-09T15:20:40.9776888Z [0., 0., 0.], 2025-09-09T15:20:40.9777094Z [0., 0., 0.]]]]) 2025-09-09T15:20:40.9777544Z PASSED 2025-09-09T15:20:40.9778188Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQATModels::test_qat_mobilenet_v2 SKIPPED 2025-09-09T15:20:40.9779136Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQATModels::test_qat_resnet18 SKIPPED 2025-09-09T15:20:40.9780068Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizeMixQATAndPTQ::test_mixing_qat_ptq PASSED 2025-09-09T15:20:40.9780943Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_add PASSED 2025-09-09T15:20:40.9781775Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_add_relu PASSED 2025-09-09T15:20:40.9782621Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_conv2d PASSED 2025-09-09T15:20:40.9783486Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_dynamic_linear PASSED 2025-09-09T15:20:40.9784363Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_maxpool2d PASSED 2025-09-09T15:20:40.9785182Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_qdq PASSED 2025-09-09T15:20:40.9786048Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_qdq_per_channel PASSED 2025-09-09T15:20:40.9786947Z test/quantization/pt2e/test_representation.py::TestPT2ERepresentation::test_static_linear PASSED 2025-09-09T15:20:40.9788276Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:40.9789952Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:40.9791612Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:40.9793258Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:40.9794968Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:40.9796729Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:40.9798389Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:40.9800054Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:40.9801810Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:40.9803548Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:40.9805252Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:40.9806901Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:40.9808553Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:40.9810222Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:40.9811877Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:40.9813514Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:40.9815174Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:40.9816843Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:40.9818486Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:40.9820139Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:40.9821796Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:40.9823457Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:40.9825106Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:40.9826759Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:40.9828403Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:49.1906143Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:49.1908256Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:49.1910359Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:49.1912403Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:20:49.1914479Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:20:49.1916654Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:20:49.1918733Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:20:49.1920692Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1922026Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:20:49.1922515Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1923833Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1925041Z graph_break [] 2025-09-09T15:20:49.1925322Z PASSED 2025-09-09T15:20:49.1926424Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1927672Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:20:49.1928146Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1929237Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1930210Z graph_break [] 2025-09-09T15:20:49.1930486Z PASSED 2025-09-09T15:20:49.1931605Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1932838Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:20:49.1933322Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1934903Z inductor [('pattern_matcher_nodes', 7), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_binary_matcher_nodes', 2), ('qlinear_weight_prepack_matcher_count', 1), ('qlinear_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_binary_lower_count', 1), ('qlinear_binary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1936583Z graph_break [] 2025-09-09T15:20:49.1936956Z PASSED 2025-09-09T15:20:49.1938060Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1939343Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:20:49.1939821Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1940897Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1941886Z graph_break [] 2025-09-09T15:20:49.1942157Z PASSED 2025-09-09T15:20:49.1943259Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1944479Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:20:49.1944950Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1946027Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1946993Z graph_break [] 2025-09-09T15:20:49.1947268Z PASSED 2025-09-09T15:20:49.1948373Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1949623Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:20:49.1950100Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1951183Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1952154Z graph_break [] 2025-09-09T15:20:49.1952424Z PASSED 2025-09-09T15:20:49.1953537Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1954776Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:20:49.1955253Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1956401Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1957365Z graph_break [] 2025-09-09T15:20:49.1957648Z PASSED 2025-09-09T15:20:49.1958755Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1959980Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:20:49.1960456Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1961528Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1962587Z graph_break [] 2025-09-09T15:20:49.1962868Z PASSED 2025-09-09T15:20:49.1964279Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1965527Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:20:49.1966001Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1967317Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1968516Z graph_break [] 2025-09-09T15:20:49.1968795Z PASSED 2025-09-09T15:20:49.1969912Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1971133Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:20:49.1980015Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:20:49.1981162Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:20:49.1982156Z graph_break [] 2025-09-09T15:20:49.1982469Z PASSED 2025-09-09T15:20:49.1983609Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:20:49.1984866Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:20:49.1985350Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0809874Z inductor [('pattern_matcher_nodes', 8), ('pattern_matcher_count', 4), ('qlinear_weight_prepack_matcher_nodes', 4), ('qlinear_binary_matcher_nodes', 2), ('qlinear_weight_prepack_matcher_count', 1), ('qlinear_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_binary_lower_count', 1), ('qlinear_binary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0811413Z graph_break [] 2025-09-09T15:21:14.0811925Z PASSED 2025-09-09T15:21:14.0813074Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0814312Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0814819Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0815905Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0816884Z graph_break [] 2025-09-09T15:21:14.0817174Z PASSED 2025-09-09T15:21:14.0818279Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0819507Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:14.0819982Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0821050Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0822488Z graph_break [] 2025-09-09T15:21:14.0822777Z PASSED 2025-09-09T15:21:14.0824048Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0825289Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0825806Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0826886Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0827885Z graph_break [] 2025-09-09T15:21:14.0828174Z PASSED 2025-09-09T15:21:14.0829284Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0830538Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:14.0831028Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0832113Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0833093Z graph_break [] 2025-09-09T15:21:14.0833374Z PASSED 2025-09-09T15:21:14.0834484Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0835716Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0836201Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0837385Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0838337Z graph_break [] 2025-09-09T15:21:14.0838628Z PASSED 2025-09-09T15:21:14.0839724Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0840932Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:21:14.0841414Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0842700Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0843887Z graph_break [] 2025-09-09T15:21:14.0844174Z PASSED 2025-09-09T15:21:14.0845261Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0846470Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0846941Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0848007Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0848967Z graph_break [] 2025-09-09T15:21:14.0849249Z PASSED 2025-09-09T15:21:14.0850465Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0851697Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:21:14.0852276Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0853599Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0854801Z graph_break [] 2025-09-09T15:21:14.0855094Z PASSED 2025-09-09T15:21:14.0856289Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0857509Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0857902Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0858743Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0859572Z graph_break [] 2025-09-09T15:21:14.0859870Z PASSED 2025-09-09T15:21:14.0861001Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0862246Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:21:14.0862726Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0864007Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0864962Z graph_break [] 2025-09-09T15:21:14.0865383Z PASSED 2025-09-09T15:21:14.0866492Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0867727Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0868209Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0869279Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0870262Z graph_break [] 2025-09-09T15:21:14.0870541Z PASSED 2025-09-09T15:21:14.0871672Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0872933Z stats [('calls_captured', 4), ('unique_graphs', 1)] 2025-09-09T15:21:14.0873414Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:14.0874508Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:14.0875497Z graph_break [] 2025-09-09T15:21:14.0875776Z PASSED 2025-09-09T15:21:14.0876992Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:14.0878393Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:14.0878875Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6033180Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6034800Z graph_break [] 2025-09-09T15:21:33.6035499Z PASSED 2025-09-09T15:21:33.6037959Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6039360Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:33.6039754Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6040796Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6041734Z graph_break [] 2025-09-09T15:21:33.6041975Z PASSED 2025-09-09T15:21:33.6043039Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6044225Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:21:33.6044668Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6045710Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6046640Z graph_break [] 2025-09-09T15:21:33.6046890Z PASSED 2025-09-09T15:21:33.6047948Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6049130Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:33.6049521Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6050536Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6051463Z graph_break [] 2025-09-09T15:21:33.6051695Z PASSED 2025-09-09T15:21:33.6052547Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6053495Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:21:33.6053885Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6054738Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6055490Z graph_break [] 2025-09-09T15:21:33.6055718Z PASSED 2025-09-09T15:21:33.6056590Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6057722Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:33.6058123Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6059038Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6059796Z graph_break [] 2025-09-09T15:21:33.6060031Z PASSED 2025-09-09T15:21:33.6060882Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6061837Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:21:33.6062214Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6063064Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6064149Z graph_break [] 2025-09-09T15:21:33.6064380Z PASSED 2025-09-09T15:21:33.6065406Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6066585Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:33.6067029Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6068066Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6068991Z graph_break [] 2025-09-09T15:21:33.6069235Z PASSED 2025-09-09T15:21:33.6070293Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_False_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:33.6071471Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:21:33.6071931Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:33.6072964Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:33.6073900Z graph_break [] 2025-09-09T15:21:33.6074142Z PASSED 2025-09-09T15:21:33.6075356Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:33.6077249Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:33.6078884Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:33.6080523Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:33.6082171Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:33.6083958Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:33.6085712Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:33.6087357Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:33.6088989Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:33.6090620Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:33.6092259Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:33.6093877Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:33.6095504Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:41.4083904Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:41.4086364Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:41.4088257Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:41.4089904Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:41.4091542Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:41.4093216Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:41.4094876Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:41.4096510Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:41.4098157Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:41.4100337Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:41.4101979Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:41.4103604Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:41.4105238Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:41.4106868Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:41.4108490Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:41.4110123Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False PASSED 2025-09-09T15:21:41.4111755Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True PASSED 2025-09-09T15:21:41.4113387Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False PASSED 2025-09-09T15:21:41.4115019Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_bfloat16_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True PASSED 2025-09-09T15:21:41.4116798Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:41.4118225Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:41.4118641Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:41.4119685Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:41.4120615Z graph_break [] 2025-09-09T15:21:41.4120870Z PASSED 2025-09-09T15:21:41.4121738Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:41.4122706Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:41.4123110Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:41.4123960Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:41.4124825Z graph_break [] 2025-09-09T15:21:41.4125058Z PASSED 2025-09-09T15:21:41.4126006Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:41.4126972Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:41.4127358Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:41.4128379Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:21:41.4129307Z graph_break [] 2025-09-09T15:21:41.4129546Z PASSED 2025-09-09T15:21:41.4130411Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:41.4131368Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:41.4131767Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:41.4132619Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:41.4133378Z graph_break [] 2025-09-09T15:21:41.4133611Z PASSED 2025-09-09T15:21:41.4134485Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:41.4135457Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:21:41.4135848Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:41.4136700Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:41.4137461Z graph_break [] 2025-09-09T15:21:41.4137692Z PASSED 2025-09-09T15:21:41.4138570Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:21:41.4139531Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:21:41.4139928Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:21:41.4140775Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:21:41.4141530Z graph_break [] 2025-09-09T15:21:41.4141765Z PASSED 2025-09-09T15:21:41.4142635Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0939690Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:22:02.0940292Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0941520Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0942422Z graph_break [] 2025-09-09T15:22:02.0942845Z PASSED 2025-09-09T15:22:02.0944220Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0945204Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:02.0945790Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0946645Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0947405Z graph_break [] 2025-09-09T15:22:02.0947652Z PASSED 2025-09-09T15:22:02.0948519Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0949493Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:02.0949877Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0950909Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0951840Z graph_break [] 2025-09-09T15:22:02.0952067Z PASSED 2025-09-09T15:22:02.0952935Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0953915Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:02.0954298Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0955145Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0955894Z graph_break [] 2025-09-09T15:22:02.0956131Z PASSED 2025-09-09T15:22:02.0957511Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0958825Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:02.0959213Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0960230Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0961167Z graph_break [] 2025-09-09T15:22:02.0961412Z PASSED 2025-09-09T15:22:02.0962412Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0963368Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:02.0963956Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0965050Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0965808Z graph_break [] 2025-09-09T15:22:02.0966054Z PASSED 2025-09-09T15:22:02.0966926Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0968188Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:02.0968581Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0969557Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0970314Z graph_break [] 2025-09-09T15:22:02.0970559Z PASSED 2025-09-09T15:22:02.0971418Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0972377Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:02.0972778Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0973613Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0974368Z graph_break [] 2025-09-09T15:22:02.0974598Z PASSED 2025-09-09T15:22:02.0975465Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0976422Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:02.0976852Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0977693Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0978442Z graph_break [] 2025-09-09T15:22:02.0978673Z PASSED 2025-09-09T15:22:02.0979527Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_False_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0980482Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:02.0980866Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0981701Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0982452Z graph_break [] 2025-09-09T15:22:02.0982678Z PASSED 2025-09-09T15:22:02.0983544Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0984508Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:22:02.0984887Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0985911Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0986843Z graph_break [] 2025-09-09T15:22:02.0987069Z PASSED 2025-09-09T15:22:02.0987932Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0988974Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:02.0989362Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0990290Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0991038Z graph_break [] 2025-09-09T15:22:02.0991270Z PASSED 2025-09-09T15:22:02.0992127Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0993080Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:22:02.0993461Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:02.0994481Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:22:02.0995413Z graph_break [] 2025-09-09T15:22:02.0995638Z PASSED 2025-09-09T15:22:02.0996732Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:02.0998179Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:02.0998742Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0289304Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0290306Z graph_break [] 2025-09-09T15:22:45.0290814Z PASSED 2025-09-09T15:22:45.0291951Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0293248Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:22:45.0293732Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0294845Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0295836Z graph_break [] 2025-09-09T15:22:45.0296126Z PASSED 2025-09-09T15:22:45.0297239Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0298482Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:45.0298967Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0300065Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0301049Z graph_break [] 2025-09-09T15:22:45.0301334Z PASSED 2025-09-09T15:22:45.0302440Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0303678Z stats [('calls_captured', 5), ('unique_graphs', 1)] 2025-09-09T15:22:45.0304158Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0305692Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0306647Z graph_break [] 2025-09-09T15:22:45.0306933Z PASSED 2025-09-09T15:22:45.0308221Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_False_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0309452Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:22:45.0309939Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0311027Z inductor [('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_nodes', 4), ('qlinear_weight_prepack_matcher_count', 1), ('pattern_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0312007Z graph_break [] 2025-09-09T15:22:45.0312296Z PASSED 2025-09-09T15:22:45.0313392Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0314633Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:45.0315119Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0316509Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0317719Z graph_break [] 2025-09-09T15:22:45.0317999Z PASSED 2025-09-09T15:22:45.0319101Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0320337Z stats [('calls_captured', 8), ('unique_graphs', 1)] 2025-09-09T15:22:45.0320814Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0321905Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0322873Z graph_break [] 2025-09-09T15:22:45.0323156Z PASSED 2025-09-09T15:22:45.0332491Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0333770Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:45.0334279Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0335622Z inductor [('pattern_matcher_nodes', 6), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0336852Z graph_break [] 2025-09-09T15:22:45.0337176Z PASSED 2025-09-09T15:22:45.0338297Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_1_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0339564Z stats [('calls_captured', 8), ('unique_graphs', 1)] 2025-09-09T15:22:45.0340092Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0341105Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0341993Z graph_break [] 2025-09-09T15:22:45.0342245Z PASSED 2025-09-09T15:22:45.0343194Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0344165Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:45.0344558Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0345402Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0346157Z graph_break [] 2025-09-09T15:22:45.0346391Z PASSED 2025-09-09T15:22:45.0347264Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_False_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0348238Z stats [('calls_captured', 8), ('unique_graphs', 1)] 2025-09-09T15:22:45.0348634Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0349479Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0350239Z graph_break [] 2025-09-09T15:22:45.0350473Z PASSED 2025-09-09T15:22:45.0351342Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0352310Z stats [('calls_captured', 6), ('unique_graphs', 1)] 2025-09-09T15:22:45.0352697Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0353551Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0354315Z graph_break [] 2025-09-09T15:22:45.0354548Z PASSED 2025-09-09T15:22:45.0355412Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_da8w8_sym_act_sym_wgt_with_int_mm_has_bias_True_float32_dynamic_True_reshape_a_True_M_32_inplace_add_True_expand_a_scale_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0356429Z stats [('calls_captured', 8), ('unique_graphs', 1)] 2025-09-09T15:22:45.0356820Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0357673Z inductor [('pattern_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:22:45.0358433Z graph_break [] 2025-09-09T15:22:45.0358669Z PASSED 2025-09-09T15:22:45.0359288Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_dynamic_qlinear_cpu stats [('calls_captured', 22), ('unique_graphs', 8)] 2025-09-09T15:22:45.0359957Z inline_call [] 2025-09-09T15:22:45.0360164Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:22:45.0360507Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:22:45.0361546Z inductor [('pattern_matcher_nodes', 10), ('qlinear_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 4), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:22:45.0362485Z graph_break [] 2025-09-09T15:22:45.0362722Z PASSED 2025-09-09T15:29:27.7839265Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_dynamic_qlinear_input_dim_exceeds_2 stats [('calls_captured', 22), ('unique_graphs', 8)] 2025-09-09T15:29:27.7841797Z inline_call [] 2025-09-09T15:29:27.7842445Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7843591Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7845341Z inductor [('pattern_matcher_nodes', 18), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 8), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7847043Z graph_break [] 2025-09-09T15:29:27.7847617Z PASSED 2025-09-09T15:29:27.7848579Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_dynamic_qlinear_qat_cpu stats [('calls_captured', 22), ('unique_graphs', 8)] 2025-09-09T15:29:27.7849616Z inline_call [] 2025-09-09T15:29:27.7849948Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7850438Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7852104Z inductor [('pattern_matcher_nodes', 10), ('qlinear_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 4), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7853777Z graph_break [] 2025-09-09T15:29:27.7854187Z PASSED 2025-09-09T15:29:27.7855152Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_linear_dynamic_fp16 stats [('calls_captured', 20), ('unique_graphs', 16)] 2025-09-09T15:29:27.7856177Z inline_call [] 2025-09-09T15:29:27.7856489Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:29:27.7856981Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:29:27.7858297Z inductor [('pattern_matcher_nodes', 15), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 5), ('extern_calls', 4), ('qlinear_weight_prepack_matcher_count', 2), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:29:27.7859421Z graph_break [] 2025-09-09T15:29:27.7859772Z PASSED 2025-09-09T15:29:27.7860701Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_linear_relu_dynamic_fp16 stats [('calls_captured', 24), ('unique_graphs', 16)] 2025-09-09T15:29:27.7861669Z inline_call [] 2025-09-09T15:29:27.7861967Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:29:27.7862436Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:29:27.7864005Z inductor [('pattern_matcher_nodes', 17), ('qlinear_weight_prepack_matcher_nodes', 14), ('pattern_matcher_count', 5), ('extern_calls', 4), ('qlinear_weight_prepack_matcher_count', 2), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:29:27.7865151Z graph_break [] 2025-09-09T15:29:27.7865529Z PASSED 2025-09-09T15:29:27.7866415Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d stats [('calls_captured', 1958), ('unique_graphs', 224)] 2025-09-09T15:29:27.7867365Z inline_call [] 2025-09-09T15:29:27.7867665Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7868120Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7869815Z inductor [('pattern_matcher_nodes', 7), ('qconv_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qconv_unary_matcher_nodes', 2), ('qconv_weight_prepack_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:29:27.7871402Z graph_break [] 2025-09-09T15:29:27.7871767Z PASSED 2025-09-09T15:29:27.7872671Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_add stats [('calls_captured', 1967), ('unique_graphs', 224)] 2025-09-09T15:29:27.7873613Z inline_call [] 2025-09-09T15:29:27.7873912Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7874711Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7877699Z inductor [('pattern_matcher_nodes', 17), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 7), ('qconv2d_binary_matcher_nodes', 4), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_nodes', 2), ('extern_calls', 2), ('dequant_promotion_matcher_count', 1), ('dequant_promotion_matcher_nodes', 1), ('qconv2d_binary_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('qconv2d_binary_lower_count', 1), ('qconv2d_binary_lower_nodes', 1)] 2025-09-09T15:29:27.7880275Z graph_break [] 2025-09-09T15:29:27.7880649Z PASSED 2025-09-09T15:29:27.7881588Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_add_relu stats [('calls_captured', 1969), ('unique_graphs', 224)] 2025-09-09T15:29:27.7882609Z inline_call [] 2025-09-09T15:29:27.7882945Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7883478Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7886221Z inductor [('pattern_matcher_nodes', 18), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 7), ('qconv2d_binary_matcher_nodes', 5), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_nodes', 2), ('extern_calls', 2), ('dequant_promotion_matcher_count', 1), ('dequant_promotion_matcher_nodes', 1), ('qconv2d_binary_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('qconv2d_binary_lower_count', 1), ('qconv2d_binary_lower_nodes', 1)] 2025-09-09T15:29:27.7888951Z graph_break [] 2025-09-09T15:29:27.7889328Z PASSED 2025-09-09T15:29:27.7890336Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_hardswish stats [('calls_captured', 1968), ('unique_graphs', 224)] 2025-09-09T15:29:27.7891495Z inline_call [] 2025-09-09T15:29:27.7891817Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7892293Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7894131Z inductor [('pattern_matcher_nodes', 24), ('qconv_unary_matcher_nodes', 14), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7895763Z graph_break [] 2025-09-09T15:29:27.7896137Z PASSED 2025-09-09T15:29:27.7897140Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_hardtanh stats [('calls_captured', 1968), ('unique_graphs', 224)] 2025-09-09T15:29:27.7898124Z inline_call [] 2025-09-09T15:29:27.7898417Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7898899Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7900907Z inductor [('pattern_matcher_nodes', 18), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv_unary_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7902593Z graph_break [] 2025-09-09T15:29:27.7902971Z PASSED 2025-09-09T15:29:27.7903935Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_relu stats [('calls_captured', 1968), ('unique_graphs', 224)] 2025-09-09T15:29:27.7904846Z inline_call [] 2025-09-09T15:29:27.7905134Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7905621Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7907725Z inductor [('pattern_matcher_nodes', 16), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_unary_matcher_nodes', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7909557Z graph_break [] 2025-09-09T15:29:27.7909928Z PASSED 2025-09-09T15:29:27.7911033Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_relu6 stats [('calls_captured', 1968), ('unique_graphs', 224)] 2025-09-09T15:29:27.7912054Z inline_call [] 2025-09-09T15:29:27.7912371Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7912855Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7914768Z inductor [('pattern_matcher_nodes', 18), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv_unary_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7916732Z graph_break [] 2025-09-09T15:29:27.7917103Z PASSED 2025-09-09T15:29:27.7918125Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qat_qconv2d_silu stats [('calls_captured', 1968), ('unique_graphs', 224)] 2025-09-09T15:29:27.7919172Z inline_call [] 2025-09-09T15:29:27.7919482Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7919992Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:29:27.7921972Z inductor [('pattern_matcher_nodes', 18), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv_unary_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:29:27.7923727Z graph_break [] 2025-09-09T15:29:27.7924075Z PASSED 2025-09-09T15:29:27.7924987Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qcat stats [('calls_captured', 26), ('unique_graphs', 8)] 2025-09-09T15:29:27.7925916Z inline_call [] 2025-09-09T15:29:27.7926215Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:29:27.7926725Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1411396Z inductor [('pattern_matcher_nodes', 18), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 7), ('qconv_unary_matcher_nodes', 4), ('qcat_matcher_nodes', 4), ('extern_calls', 4), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('fxgraph_cache_bypass', 1), ('qcat_matcher_count', 1)] 2025-09-09T15:31:24.1422714Z graph_break [] 2025-09-09T15:31:24.1423197Z PASSED 2025-09-09T15:31:24.1423977Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv1d_relu_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:31:24.1424814Z inline_call [] 2025-09-09T15:31:24.1425091Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1425498Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1427085Z inductor [('pattern_matcher_nodes', 13), ('qconv_weight_prepack_matcher_nodes', 6), ('pattern_matcher_count', 6), ('qconv_unary_matcher_nodes', 5), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:31:24.1428510Z graph_break [] 2025-09-09T15:31:24.1428787Z PASSED 2025-09-09T15:31:24.1429530Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_2 stats [('calls_captured', 13), ('unique_graphs', 8)] 2025-09-09T15:31:24.1430336Z inline_call [] 2025-09-09T15:31:24.1430600Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1431005Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1432638Z inductor [('pattern_matcher_nodes', 5), ('qconv_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qconv_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:31:24.1433868Z graph_break [] 2025-09-09T15:31:24.1434147Z PASSED 2025-09-09T15:31:24.1435076Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_3 stats [('calls_captured', 29), ('unique_graphs', 8)] 2025-09-09T15:31:24.1435886Z inline_call [] 2025-09-09T15:31:24.1436142Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1436636Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1439162Z inductor [('pattern_matcher_nodes', 18), ('pattern_matcher_count', 8), ('qconv_weight_prepack_matcher_nodes', 7), ('qcat_matcher_nodes', 4), ('extern_calls', 4), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_nodes', 2), ('qconv2d_binary_matcher_nodes', 2), ('dequant_promotion_matcher_count', 1), ('dequant_promotion_matcher_nodes', 1), ('qconv_unary_matcher_count', 1), ('qconv2d_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv2d_binary_lower_count', 1), ('qconv2d_binary_lower_nodes', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('qcat_matcher_count', 1)] 2025-09-09T15:31:24.1441580Z graph_break [] 2025-09-09T15:31:24.1441856Z PASSED 2025-09-09T15:31:24.1442688Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_broadcast_shapes_cpu stats [('calls_captured', 15), ('unique_graphs', 8)] 2025-09-09T15:31:24.1443594Z inline_call [] 2025-09-09T15:31:24.1443835Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1444240Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1445514Z inductor [('pattern_matcher_nodes', 5), ('qconv_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 2), ('qconv_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:31:24.1446693Z graph_break [] 2025-09-09T15:31:24.1446971Z PASSED 2025-09-09T15:31:24.1447547Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_cpu inline_call [] 2025-09-09T15:31:24.1448277Z stats [('calls_captured', 24), ('unique_graphs', 8)] 2025-09-09T15:31:24.1448656Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1449062Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1450648Z inductor [('pattern_matcher_nodes', 16), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv2d_binary_matcher_nodes', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv2d_binary_matcher_count', 2), ('qconv2d_binary_lower_count', 2), ('qconv2d_binary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:31:24.1452122Z graph_break [] 2025-09-09T15:31:24.1452403Z PASSED 2025-09-09T15:31:24.1453168Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_int8_mixed_bf16 SKIPPED 2025-09-09T15:31:24.1454265Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_relu_cpu inline_call [] 2025-09-09T15:31:24.1455009Z stats [('calls_captured', 28), ('unique_graphs', 8)] 2025-09-09T15:31:24.1455392Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1455789Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1457373Z inductor [('pattern_matcher_nodes', 18), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv2d_binary_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv2d_binary_matcher_count', 2), ('qconv2d_binary_lower_count', 2), ('qconv2d_binary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:31:24.1458862Z graph_break [] 2025-09-09T15:31:24.1459122Z PASSED 2025-09-09T15:31:24.1459862Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_relu_int8_mixed_bf16 SKIPPED 2025-09-09T15:31:24.1460824Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_cpu stats [('calls_captured', 21), ('unique_graphs', 8)] 2025-09-09T15:31:24.1461525Z inline_call [] 2025-09-09T15:31:24.1461735Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1462055Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1463223Z inductor [('pattern_matcher_nodes', 19), ('qconv_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 8), ('qconv_unary_matcher_nodes', 4), ('qconv_weight_prepack_matcher_count', 3), ('qconv_unary_lower_count', 3), ('qconv_unary_lower_nodes', 3), ('extern_calls', 3), ('qconv_unary_matcher_count', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:31:24.1464484Z graph_break [] 2025-09-09T15:31:24.1464713Z PASSED 2025-09-09T15:31:24.1465360Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_dequant_promotion_cpu stats [('calls_captured', 24), ('unique_graphs', 8)] 2025-09-09T15:31:24.1466046Z inline_call [] 2025-09-09T15:31:24.1466248Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1466571Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1468367Z inductor [('pattern_matcher_nodes', 22), ('qconv_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 10), ('qconv_unary_matcher_nodes', 4), ('qconv_weight_prepack_matcher_count', 3), ('extern_calls', 3), ('qconv_unary_matcher_count', 2), ('qconv2d_binary_matcher_nodes', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('dequant_promotion_matcher_count', 1), ('dequant_promotion_matcher_nodes', 1), ('qconv2d_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv2d_binary_lower_count', 1), ('qconv2d_binary_lower_nodes', 1)] 2025-09-09T15:31:24.1470062Z graph_break [] 2025-09-09T15:31:24.1470292Z PASSED 2025-09-09T15:31:24.1470905Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardswish_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:31:24.1471578Z inline_call [] 2025-09-09T15:31:24.1471779Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1472108Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1473273Z inductor [('pattern_matcher_nodes', 23), ('qconv_unary_matcher_nodes', 13), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:31:24.1474343Z graph_break [] 2025-09-09T15:31:24.1474573Z PASSED 2025-09-09T15:31:24.1475231Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardswish_int8_mixed_bf16_cpu SKIPPED 2025-09-09T15:31:24.1476252Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardtanh_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:31:24.1476991Z inline_call [] 2025-09-09T15:31:24.1477201Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1477529Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:31:24.1478710Z inductor [('pattern_matcher_nodes', 17), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv_unary_matcher_nodes', 7), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:31:24.1479783Z graph_break [] 2025-09-09T15:31:24.1480009Z PASSED 2025-09-09T15:31:24.1480666Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_hardtanh_int8_mixed_bf16_cpu SKIPPED 2025-09-09T15:31:24.1481815Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_int8_mixed_bf16 SKIPPED 2025-09-09T15:31:24.1482768Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_relu6_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:31:24.1483415Z inline_call [] 2025-09-09T15:31:24.1483726Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:31:24.1484063Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:35:48.5180873Z inductor [('pattern_matcher_nodes', 17), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv_unary_matcher_nodes', 7), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:35:48.5182350Z graph_break [] 2025-09-09T15:35:48.5182851Z PASSED 2025-09-09T15:35:48.5183629Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_relu_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:35:48.5184485Z inline_call [] 2025-09-09T15:35:48.5184733Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:35:48.5185154Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:35:48.5186696Z inductor [('pattern_matcher_nodes', 15), ('qconv_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qconv_unary_matcher_nodes', 5), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:35:48.5188109Z graph_break [] 2025-09-09T15:35:48.5188393Z PASSED 2025-09-09T15:35:48.5189196Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_relu_int8_mixed_bf16_xpu SKIPPED 2025-09-09T15:35:48.5190461Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_silu_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:35:48.5191285Z inline_call [] 2025-09-09T15:35:48.5191534Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:35:48.5191960Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:35:48.5193488Z inductor [('pattern_matcher_nodes', 17), ('qconv_weight_prepack_matcher_nodes', 8), ('qconv_unary_matcher_nodes', 7), ('pattern_matcher_count', 6), ('qconv_weight_prepack_matcher_count', 2), ('qconv_unary_matcher_count', 2), ('qconv_unary_lower_count', 2), ('qconv_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:35:48.5194900Z graph_break [] 2025-09-09T15:35:48.5195178Z PASSED 2025-09-09T15:35:48.5195980Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_silu_int8_mixed_bf16_cpu SKIPPED 2025-09-09T15:35:48.5197388Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_with_concat_cpu stats [('calls_captured', 32), ('unique_graphs', 8)] 2025-09-09T15:35:48.5198252Z inline_call [] 2025-09-09T15:35:48.5198501Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:35:48.5198912Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:35:48.5200879Z inductor [('pattern_matcher_nodes', 30), ('pattern_matcher_count', 14), ('qconv_weight_prepack_matcher_nodes', 13), ('qconv_unary_matcher_nodes', 6), ('extern_calls', 6), ('qcat_matcher_nodes', 5), ('qconv_weight_prepack_matcher_count', 4), ('qconv_unary_lower_count', 4), ('qconv_unary_lower_nodes', 4), ('qconv_unary_matcher_count', 3), ('dequant_promotion_matcher_count', 2), ('dequant_promotion_matcher_nodes', 2), ('fxgraph_cache_bypass', 1), ('qcat_matcher_count', 1)] 2025-09-09T15:35:48.5202733Z graph_break [] 2025-09-09T15:35:48.5203013Z PASSED 2025-09-09T15:35:48.5203728Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qflatten stats [('calls_captured', 27), ('unique_graphs', 8)] 2025-09-09T15:35:48.5206747Z inline_call [] 2025-09-09T15:35:48.5206993Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:35:48.5207397Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:35:48.5209347Z inductor [('pattern_matcher_nodes', 12), ('pattern_matcher_count', 5), ('qconv_weight_prepack_matcher_nodes', 4), ('qconv_unary_matcher_nodes', 3), ('qreshape_matcher_nodes', 3), ('qconv_weight_prepack_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('qreshape_matcher_count', 1), ('extern_calls', 1)] 2025-09-09T15:35:48.5210981Z graph_break [] 2025-09-09T15:35:48.5211268Z PASSED 2025-09-09T15:35:48.5212043Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_False_is_dynamic_False inline_call [] 2025-09-09T15:35:48.5212950Z stats [('calls_captured', 56), ('unique_graphs', 16)] 2025-09-09T15:35:48.5213334Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:35:48.5213745Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:35:48.5216021Z inductor [('pattern_matcher_nodes', 102), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 10), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:35:48.5218163Z graph_break [] 2025-09-09T15:35:48.5218431Z PASSED 2025-09-09T15:35:48.5219211Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_False_is_dynamic_True inline_call [] 2025-09-09T15:35:48.5220128Z stats [('calls_captured', 60), ('unique_graphs', 16)] 2025-09-09T15:35:48.5220515Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:35:48.5220919Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:35:48.5223217Z inductor [('pattern_matcher_nodes', 101), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 9), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:35:48.5225337Z graph_break [] 2025-09-09T15:35:48.5225571Z PASSED 2025-09-09T15:35:48.5226175Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_True_is_dynamic_False inline_call [] 2025-09-09T15:35:48.5226897Z stats [('calls_captured', 56), ('unique_graphs', 16)] 2025-09-09T15:35:48.5227204Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:35:48.5227530Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:35:48.5229286Z inductor [('pattern_matcher_nodes', 102), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 10), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:35:48.5230935Z graph_break [] 2025-09-09T15:35:48.5231162Z PASSED 2025-09-09T15:35:48.5231770Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_False_is_qat_True_is_dynamic_True inline_call [] 2025-09-09T15:35:48.5232578Z stats [('calls_captured', 60), ('unique_graphs', 16)] 2025-09-09T15:35:48.5232895Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:35:48.5233212Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:35:48.5235044Z inductor [('pattern_matcher_nodes', 101), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 9), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:35:48.5236767Z graph_break [] 2025-09-09T15:35:48.5236995Z PASSED 2025-09-09T15:35:48.5237611Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_False_is_dynamic_False inline_call [] 2025-09-09T15:35:48.5238327Z stats [('calls_captured', 64), ('unique_graphs', 16)] 2025-09-09T15:35:48.5238642Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:35:48.5238969Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:35:48.5240718Z inductor [('pattern_matcher_nodes', 106), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 14), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:35:48.5242371Z graph_break [] 2025-09-09T15:35:48.5242601Z PASSED 2025-09-09T15:35:48.5243202Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_False_is_dynamic_True inline_call [] 2025-09-09T15:35:48.5243915Z stats [('calls_captured', 68), ('unique_graphs', 16)] 2025-09-09T15:35:48.5244222Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:35:48.5244551Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:38:06.1020208Z inductor [('pattern_matcher_nodes', 105), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 13), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:38:06.1022473Z graph_break [] 2025-09-09T15:38:06.1022999Z PASSED 2025-09-09T15:38:06.1023805Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_True_is_dynamic_False inline_call [] 2025-09-09T15:38:06.1024733Z stats [('calls_captured', 64), ('unique_graphs', 16)] 2025-09-09T15:38:06.1025126Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:38:06.1025556Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:38:06.1027935Z inductor [('pattern_matcher_nodes', 106), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 14), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:38:06.1030429Z graph_break [] 2025-09-09T15:38:06.1030716Z PASSED 2025-09-09T15:38:06.1031493Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_cpu_use_relu_True_is_qat_True_is_dynamic_True inline_call [] 2025-09-09T15:38:06.1032402Z stats [('calls_captured', 68), ('unique_graphs', 16)] 2025-09-09T15:38:06.1032976Z frames [('total', 2), ('ok', 2)] 2025-09-09T15:38:06.1033389Z aot_autograd [('total', 2), ('autograd_cache_bypass', 2), ('ok', 2)] 2025-09-09T15:38:06.1035730Z inductor [('pattern_matcher_nodes', 105), ('pattern_matcher_count', 48), ('qlinear_weight_prepack_matcher_nodes', 48), ('qlinear_binary_matcher_nodes', 13), ('dequant_promotion_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_count', 8), ('extern_calls', 8), ('removed_pointless_view_pair', 4), ('dequant_promotion_matcher_count', 4), ('qlinear_binary_matcher_count', 4), ('qlinear_unary_lower_count', 4), ('qlinear_unary_lower_nodes', 4), ('qlinear_binary_lower_count', 4), ('qlinear_binary_lower_nodes', 4), ('fxgraph_cache_bypass', 2)] 2025-09-09T15:38:06.1038094Z graph_break [] 2025-09-09T15:38:06.1038380Z PASSED 2025-09-09T15:38:06.1039394Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_False_is_dynamic_False SKIPPED 2025-09-09T15:38:06.1041071Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_False_is_dynamic_True SKIPPED 2025-09-09T15:38:06.1042724Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_True_is_dynamic_False SKIPPED 2025-09-09T15:38:06.1044375Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_False_is_qat_True_is_dynamic_True SKIPPED 2025-09-09T15:38:06.1046026Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_False_is_dynamic_False SKIPPED 2025-09-09T15:38:06.1047679Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_False_is_dynamic_True SKIPPED 2025-09-09T15:38:06.1049329Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_True_is_dynamic_False SKIPPED 2025-09-09T15:38:06.1050969Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_add_int8_mixed_bf16_use_relu_True_is_qat_True_is_dynamic_True SKIPPED 2025-09-09T15:38:06.1052364Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_cpu stats [('calls_captured', 16), ('unique_graphs', 8)] 2025-09-09T15:38:06.1053174Z inline_call [] 2025-09-09T15:38:06.1053421Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:38:06.1053831Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:38:06.1055441Z inductor [('pattern_matcher_nodes', 12), ('qlinear_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 5), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_matcher_nodes', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('qlinear_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:38:06.1056934Z graph_break [] 2025-09-09T15:38:06.1057211Z PASSED 2025-09-09T15:38:06.1058033Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_cpu stats [('calls_captured', 22), ('unique_graphs', 8)] 2025-09-09T15:38:06.1058938Z inline_call [] 2025-09-09T15:38:06.1059192Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:38:06.1059594Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:38:06.1062133Z inductor [('pattern_matcher_nodes', 20), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 9), ('qlinear_weight_prepack_matcher_count', 3), ('extern_calls', 3), ('qlinear_unary_matcher_nodes', 2), ('qlinear_binary_matcher_nodes', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('dequant_promotion_matcher_count', 1), ('dequant_promotion_matcher_nodes', 1), ('qlinear_unary_matcher_count', 1), ('qlinear_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_binary_lower_count', 1), ('qlinear_binary_lower_nodes', 1)] 2025-09-09T15:38:06.1064932Z graph_break [] 2025-09-09T15:38:06.1065217Z PASSED 2025-09-09T15:38:06.1066128Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_cpu_input_dim_exceeds_2 stats [('calls_captured', 22), ('unique_graphs', 8)] 2025-09-09T15:38:06.1067092Z inline_call [] 2025-09-09T15:38:06.1067341Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:38:06.1067738Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:38:06.1070194Z inductor [('pattern_matcher_nodes', 33), ('qlinear_weight_prepack_matcher_nodes', 18), ('pattern_matcher_count', 15), ('qlinear_weight_prepack_matcher_count', 3), ('extern_calls', 3), ('dequant_promotion_matcher_nodes', 2), ('qlinear_unary_matcher_nodes', 2), ('qlinear_binary_matcher_nodes', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('dequant_promotion_matcher_count', 1), ('qlinear_unary_matcher_count', 1), ('qlinear_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_binary_lower_count', 1), ('qlinear_binary_lower_nodes', 1)] 2025-09-09T15:38:06.1072157Z graph_break [] 2025-09-09T15:38:06.1072389Z PASSED 2025-09-09T15:38:06.1073073Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_dynamic_cpu stats [('calls_captured', 27), ('unique_graphs', 8)] 2025-09-09T15:38:06.1073803Z inline_call [] 2025-09-09T15:38:06.1074008Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:38:06.1074346Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:38:06.1076015Z inductor [('pattern_matcher_nodes', 18), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 8), ('qlinear_weight_prepack_matcher_count', 3), ('extern_calls', 3), ('qlinear_binary_matcher_nodes', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('dequant_promotion_matcher_count', 1), ('dequant_promotion_matcher_nodes', 1), ('qlinear_binary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_binary_lower_count', 1), ('qlinear_binary_lower_nodes', 1)] 2025-09-09T15:38:06.1077666Z graph_break [] 2025-09-09T15:38:06.1077903Z PASSED 2025-09-09T15:38:06.1078578Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_int8_mixed_bf16 SKIPPED 2025-09-09T15:38:06.1079762Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_dequant_promotion_int8_mixed_bf16_input_dim_exceeds_2 SKIPPED 2025-09-09T15:38:06.1080856Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_gelu_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:38:06.1081501Z inline_call [] 2025-09-09T15:38:06.1081708Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:38:06.1082033Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:38:06.1083281Z inductor [('pattern_matcher_nodes', 31), ('qlinear_unary_matcher_nodes', 21), ('qlinear_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_matcher_count', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:38:06.1084419Z graph_break [] 2025-09-09T15:38:06.1084646Z PASSED 2025-09-09T15:38:06.1085275Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_gelu_int8_mixed_bf16 SKIPPED 2025-09-09T15:38:06.1086279Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_input_dim_exceeds_2 stats [('calls_captured', 16), ('unique_graphs', 8)] 2025-09-09T15:38:06.1087118Z inline_call [] 2025-09-09T15:38:06.1087320Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6848238Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6854021Z inductor [('pattern_matcher_nodes', 20), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 9), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_matcher_nodes', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('qlinear_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:39:01.6855249Z graph_break [] 2025-09-09T15:39:01.6855671Z PASSED 2025-09-09T15:39:01.6856410Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_input_dim_exceeds_2_and_not_contiguous stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:39:01.6857166Z inline_call [] 2025-09-09T15:39:01.6857384Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6857716Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6858952Z inductor [('pattern_matcher_nodes', 20), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 9), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_matcher_nodes', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('qlinear_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:39:01.6860085Z graph_break [] 2025-09-09T15:39:01.6860319Z PASSED 2025-09-09T15:39:01.6860934Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_int8_mixed_bf16 SKIPPED 2025-09-09T15:39:01.6861987Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_int8_mixed_bf16_input_dim_exceeds_2 SKIPPED 2025-09-09T15:39:01.6863152Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_int8_mixed_bf16_input_dim_exceeds_2_and_not_contiguous SKIPPED 2025-09-09T15:39:01.6864883Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_mul_cpu stats [('calls_captured', 17), ('unique_graphs', 8)] 2025-09-09T15:39:01.6865607Z inline_call [] 2025-09-09T15:39:01.6865820Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6866157Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6867375Z inductor [('pattern_matcher_nodes', 7), ('qlinear_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qlinear_unary_matcher_nodes', 2), ('qlinear_weight_prepack_matcher_count', 1), ('qlinear_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:39:01.6868500Z graph_break [] 2025-09-09T15:39:01.6868749Z PASSED 2025-09-09T15:39:01.6869341Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_cpu stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:39:01.6869993Z inline_call [] 2025-09-09T15:39:01.6870197Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6870534Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6871773Z inductor [('pattern_matcher_nodes', 15), ('qlinear_weight_prepack_matcher_nodes', 8), ('pattern_matcher_count', 6), ('qlinear_unary_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_matcher_count', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:39:01.6872893Z graph_break [] 2025-09-09T15:39:01.6873130Z PASSED 2025-09-09T15:39:01.6873772Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_input_dim_exceeds_2 stats [('calls_captured', 20), ('unique_graphs', 8)] 2025-09-09T15:39:01.6874688Z inline_call [] 2025-09-09T15:39:01.6874895Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6875228Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6876699Z inductor [('pattern_matcher_nodes', 23), ('qlinear_weight_prepack_matcher_nodes', 12), ('pattern_matcher_count', 10), ('qlinear_unary_matcher_nodes', 5), ('qlinear_weight_prepack_matcher_count', 2), ('qlinear_unary_matcher_count', 2), ('qlinear_unary_lower_count', 2), ('qlinear_unary_lower_nodes', 2), ('extern_calls', 2), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:39:01.6877824Z graph_break [] 2025-09-09T15:39:01.6878062Z PASSED 2025-09-09T15:39:01.6878677Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_int8_mixed_bf16 SKIPPED 2025-09-09T15:39:01.6879752Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qlinear_relu_int8_mixed_bf16_input_dim_exceeds_2 SKIPPED 2025-09-09T15:39:01.6880769Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qmaxpool2d stats [('calls_captured', 19), ('unique_graphs', 8)] 2025-09-09T15:39:01.6881398Z inline_call [] 2025-09-09T15:39:01.6881607Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6881931Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6883292Z inductor [('pattern_matcher_nodes', 12), ('qconv_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 4), ('qmaxpool2d_matcher_nodes', 4), ('qconv_unary_matcher_nodes', 3), ('extern_calls', 3), ('qconv_weight_prepack_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('qmaxpool2d_matcher_count', 1)] 2025-09-09T15:39:01.6884552Z graph_break [] 2025-09-09T15:39:01.6884776Z PASSED 2025-09-09T15:39:01.6885617Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_False_dynamic_False PASSED 2025-09-09T15:39:01.6887003Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_False_dynamic_True PASSED 2025-09-09T15:39:01.6888369Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_True_dynamic_False PASSED 2025-09-09T15:39:01.6889736Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_bfloat16_per_channel_quant_True_dynamic_True PASSED 2025-09-09T15:39:01.6891015Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_False_dynamic_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6891856Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:39:01.6892250Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6893096Z inductor [('pattern_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_nodes', 6), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:39:01.6893907Z graph_break [] 2025-09-09T15:39:01.6894130Z PASSED 2025-09-09T15:39:01.6894875Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_False_dynamic_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6895722Z stats [('calls_captured', 10), ('unique_graphs', 1)] 2025-09-09T15:39:01.6896109Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6897131Z inductor [('pattern_matcher_nodes', 10), ('qlinear_weight_prepack_matcher_nodes', 7), ('pattern_matcher_count', 4), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:39:01.6898147Z graph_break [] 2025-09-09T15:39:01.6898387Z PASSED 2025-09-09T15:39:01.6899133Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_True_dynamic_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6900047Z stats [('calls_captured', 7), ('unique_graphs', 1)] 2025-09-09T15:39:01.6900443Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6901284Z inductor [('pattern_matcher_nodes', 8), ('qlinear_weight_prepack_matcher_nodes', 6), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:39:01.6902038Z graph_break [] 2025-09-09T15:39:01.6902277Z PASSED 2025-09-09T15:39:01.6903006Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_False_float32_per_channel_quant_True_dynamic_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:39:01.6903900Z stats [('calls_captured', 10), ('unique_graphs', 1)] 2025-09-09T15:39:01.6904293Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:39:01.6905140Z inductor [('pattern_matcher_nodes', 9), ('qlinear_weight_prepack_matcher_nodes', 7), ('pattern_matcher_count', 3), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:39:01.6905890Z graph_break [] 2025-09-09T15:39:01.6906117Z PASSED 2025-09-09T15:39:01.6906944Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_False_dynamic_False PASSED 2025-09-09T15:39:01.6908305Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_False_dynamic_True PASSED 2025-09-09T15:39:01.6909665Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_True_dynamic_False PASSED 2025-09-09T15:39:01.6911023Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_bfloat16_per_channel_quant_True_dynamic_True PASSED 2025-09-09T15:40:35.8063212Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_False_dynamic_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8064491Z stats [('calls_captured', 10), ('unique_graphs', 1)] 2025-09-09T15:40:35.8064889Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8065854Z inductor [('pattern_matcher_nodes', 12), ('qlinear_weight_prepack_matcher_nodes', 7), ('pattern_matcher_count', 5), ('removed_pointless_view_pair', 1), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:40:35.8066747Z graph_break [] 2025-09-09T15:40:35.8067163Z PASSED 2025-09-09T15:40:35.8067906Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_False_dynamic_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8068741Z stats [('calls_captured', 14), ('unique_graphs', 1)] 2025-09-09T15:40:35.8069131Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8070256Z inductor [('pattern_matcher_nodes', 13), ('qlinear_weight_prepack_matcher_nodes', 7), ('pattern_matcher_count', 6), ('removed_pointless_view_pair', 1), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:40:35.8071276Z graph_break [] 2025-09-09T15:40:35.8071507Z PASSED 2025-09-09T15:40:35.8072239Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_True_dynamic_False frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8073553Z stats [('calls_captured', 10), ('unique_graphs', 1)] 2025-09-09T15:40:35.8073945Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8075056Z inductor [('pattern_matcher_nodes', 12), ('qlinear_weight_prepack_matcher_nodes', 7), ('pattern_matcher_count', 5), ('removed_pointless_view_pair', 1), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:40:35.8075972Z graph_break [] 2025-09-09T15:40:35.8076199Z PASSED 2025-09-09T15:40:35.8077030Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_smooth_quant_with_int_mm_has_bias_True_float32_per_channel_quant_True_dynamic_True frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8077862Z stats [('calls_captured', 14), ('unique_graphs', 1)] 2025-09-09T15:40:35.8078251Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8079206Z inductor [('pattern_matcher_nodes', 12), ('qlinear_weight_prepack_matcher_nodes', 7), ('pattern_matcher_count', 5), ('removed_pointless_view_pair', 1), ('qlinear_weight_prepack_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('extern_calls', 1)] 2025-09-09T15:40:35.8080056Z graph_break [] 2025-09-09T15:40:35.8080290Z PASSED 2025-09-09T15:40:35.8080803Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_q_attention_block inline_call [] 2025-09-09T15:40:35.8081413Z stats [('calls_captured', 49), ('unique_graphs', 8)] 2025-09-09T15:40:35.8081726Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8082047Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8083507Z inductor [('pattern_matcher_nodes', 51), ('pattern_matcher_count', 29), ('qlinear_weight_prepack_matcher_nodes', 18), ('qlinear_unary_matcher_nodes', 6), ('extern_calls', 5), ('qlinear_weight_prepack_matcher_count', 3), ('qlinear_unary_matcher_count', 3), ('qlinear_unary_lower_count', 3), ('qlinear_unary_lower_nodes', 3), ('dequant_promotion_matcher_nodes', 2), ('dequant_promotion_matcher_count', 1), ('fxgraph_cache_bypass', 1)] 2025-09-09T15:40:35.8084866Z graph_break [] 2025-09-09T15:40:35.8085156Z aten_mm_info [('aten.bmm_32_384_384_64', 1), ('aten.bmm_32_384_64_384', 1)] 2025-09-09T15:40:35.8085541Z PASSED 2025-09-09T15:40:35.8086166Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_qat_bn_conv2d stats [('calls_captured', 1960), ('unique_graphs', 224)] 2025-09-09T15:40:35.8086844Z inline_call [] 2025-09-09T15:40:35.8087046Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8087380Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8088555Z inductor [('pattern_matcher_nodes', 7), ('qconv_weight_prepack_matcher_nodes', 4), ('pattern_matcher_count', 3), ('qconv_unary_matcher_nodes', 2), ('qconv_weight_prepack_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('extern_calls', 1)] 2025-09-09T15:40:35.8089626Z graph_break [] 2025-09-09T15:40:35.8089857Z PASSED 2025-09-09T15:40:35.8090561Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_qconv2d_maxpool2d_linear_dynamic_cpu stats [('calls_captured', 30), ('unique_graphs', 8)] 2025-09-09T15:40:35.8091313Z inline_call [] 2025-09-09T15:40:35.8091518Z frames [('total', 1), ('ok', 1)] 2025-09-09T15:40:35.8091840Z aot_autograd [('total', 1), ('autograd_cache_bypass', 1), ('ok', 1)] 2025-09-09T15:40:35.8093815Z inductor [('pattern_matcher_nodes', 21), ('pattern_matcher_count', 8), ('qlinear_weight_prepack_matcher_nodes', 4), ('qconv_weight_prepack_matcher_nodes', 4), ('qmaxpool2d_matcher_nodes', 4), ('extern_calls', 4), ('qconv_unary_matcher_nodes', 3), ('qreshape_matcher_nodes', 3), ('qlinear_weight_prepack_matcher_count', 1), ('qconv_weight_prepack_matcher_count', 1), ('qconv_unary_matcher_count', 1), ('fxgraph_cache_bypass', 1), ('qconv_unary_lower_count', 1), ('qconv_unary_lower_nodes', 1), ('qmaxpool2d_matcher_count', 1), ('qreshape_matcher_count', 1), ('qlinear_unary_lower_count', 1), ('qlinear_unary_lower_nodes', 1)] 2025-09-09T15:40:35.8095799Z graph_break [] 2025-09-09T15:40:35.8096097Z PASSED 2025-09-09T15:40:35.8096787Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_adaptive_avg_pool2d_recipe PASSED 2025-09-09T15:40:35.8097861Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_annotate_mul_tensor PASSED 2025-09-09T15:40:35.8098875Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_attention_block PASSED 2025-09-09T15:40:35.8099884Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_avg_pool2d_recipe PASSED 2025-09-09T15:40:35.8100871Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_cat_recipe PASSED 2025-09-09T15:40:35.8101879Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_cat_recipe_same_inputs PASSED 2025-09-09T15:40:35.8102946Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_cat_recipe_single_input PASSED 2025-09-09T15:40:35.8103933Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d PASSED 2025-09-09T15:40:35.8104889Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_binary PASSED 2025-09-09T15:40:35.8105873Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_binary2 PASSED 2025-09-09T15:40:35.8106877Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_binary_unary PASSED 2025-09-09T15:40:35.8107947Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_serials_binary_unary PASSED 2025-09-09T15:40:35.8108968Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_conv2d_unary PASSED 2025-09-09T15:40:35.8109972Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_dynamic_quant_linear PASSED 2025-09-09T15:40:35.8111006Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_conv2d_recipe PASSED 2025-09-09T15:40:35.8112031Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_linear_recipe PASSED 2025-09-09T15:40:35.8113076Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_maxpool2d_recipe PASSED 2025-09-09T15:40:35.8114090Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_flatten_recipe PASSED 2025-09-09T15:40:35.8115089Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_flatten_recipe2 PASSED 2025-09-09T15:40:35.8116042Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear PASSED 2025-09-09T15:40:35.8117060Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary PASSED 2025-09-09T15:40:35.8118043Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary2 PASSED 2025-09-09T15:40:35.8119054Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_dynamic PASSED 2025-09-09T15:40:35.8120110Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_dynamic_qat PASSED 2025-09-09T15:40:35.8121149Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_qat PASSED 2025-09-09T15:40:35.8122258Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary PASSED 2025-09-09T15:40:35.8123402Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_dynamic PASSED 2025-09-09T15:45:43.9248488Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_dynamic_qat PASSED 2025-09-09T15:45:43.9249679Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_qat PASSED 2025-09-09T15:45:43.9250761Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_binary_unary_serials PASSED 2025-09-09T15:45:43.9251828Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_dynamic_fp16 PASSED 2025-09-09T15:45:43.9252876Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary PASSED 2025-09-09T15:45:43.9253884Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary_dynamic PASSED 2025-09-09T15:45:43.9254942Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary_dynamic_qat PASSED 2025-09-09T15:45:43.9255984Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_linear_unary_qat PASSED 2025-09-09T15:45:43.9256984Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_lowering_to_x86 SKIPPED 2025-09-09T15:45:43.9257992Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_maxpool2d_recipe PASSED 2025-09-09T15:45:43.9258974Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d PASSED 2025-09-09T15:45:43.9259954Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_binary PASSED 2025-09-09T15:45:43.9260973Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_binary2 PASSED 2025-09-09T15:45:43.9262021Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_binary_unary PASSED 2025-09-09T15:45:43.9263056Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_conv2d_unary PASSED 2025-09-09T15:45:43.9264323Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_qat_dynamic_quant_linear PASSED 2025-09-09T15:45:43.9265438Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_case1 PASSED 2025-09-09T15:45:43.9266588Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_case2 PASSED 2025-09-09T15:45:43.9267799Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_with_mixed_configs PASSED 2025-09-09T15:45:43.9268942Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_qconfig PASSED 2025-09-09T15:45:43.9270069Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_qconfig_for_dynamic_quant PASSED 2025-09-09T15:45:43.9271250Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_qconfig_with_underscores PASSED 2025-09-09T15:45:43.9272405Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_with_mixed_configs PASSED 2025-09-09T15:45:43.9273429Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_bmm SKIPPED 2025-09-09T15:45:43.9274981Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity0_sizes0 SKIPPED 2025-09-09T15:45:43.9276337Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity0_sizes1 SKIPPED 2025-09-09T15:45:43.9277419Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity0_sizes2 SKIPPED 2025-09-09T15:45:43.9278506Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity1_sizes0 SKIPPED 2025-09-09T15:45:43.9279588Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity1_sizes1 SKIPPED 2025-09-09T15:45:43.9280665Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_cat_granularity1_sizes2 SKIPPED 2025-09-09T15:45:43.9281771Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_expected_gpu_kernel_fbgemm SKIPPED 2025-09-09T15:45:43.9283250Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9285024Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9286798Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9288627Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9299326Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9301130Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9302916Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9304703Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9306503Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9308293Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9310091Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9311875Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9313860Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9315634Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9317476Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9319258Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9321060Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9322844Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9516013Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9517914Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9519718Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9521524Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9523306Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9525097Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9526934Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9528764Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9530603Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9532452Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9534606Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9536442Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9538259Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9540082Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9541921Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9543760Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9545594Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9547420Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9549249Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9551060Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9552867Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9554702Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9556627Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9558441Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9560253Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9562074Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9564258Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9566113Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9567957Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9569793Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_bfloat16_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9571599Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9573378Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9575167Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9576959Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9578765Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9580562Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9794214Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9796015Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9797897Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9799696Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9801476Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9803268Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9805383Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9807155Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9808945Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9810738Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9812529Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9814312Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9816089Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9817855Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9819640Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9821432Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9823212Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9824987Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_dynamic_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9826799Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9828621Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9830435Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9832273Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9834197Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9836087Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9837979Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9839793Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9841613Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9843450Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9845274Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9847086Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_False_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:45:43.9848902Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:45:43.9850710Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:45:43.9852519Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:45:43.9854333Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:45:43.9856153Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:45:43.9857956Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity0_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:46:22.5178666Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes0 SKIPPED 2025-09-09T15:46:22.5181101Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_AUTO_sizes1 SKIPPED 2025-09-09T15:46:22.5183967Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes0 SKIPPED 2025-09-09T15:46:22.5187220Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_FBGEMM_sizes1 SKIPPED 2025-09-09T15:46:22.5189937Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes0 SKIPPED 2025-09-09T15:46:22.5192362Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_fp8_linear_variants_float32_mode_weight-only_compile_True_granularity1_kernel_preference_KernelPreference_TORCH_sizes1 SKIPPED 2025-09-09T15:46:22.5194648Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity0_sizes0 SKIPPED 2025-09-09T15:46:22.5196495Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity0_sizes1 SKIPPED 2025-09-09T15:46:22.5198272Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity1_sizes0 SKIPPED 2025-09-09T15:46:22.5200031Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_kernel_preference_numerical_equivalence_granularity1_sizes1 SKIPPED 2025-09-09T15:46:22.5201613Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_moe_weight_reshape_ops SKIPPED 2025-09-09T15:46:22.5206969Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_and_copy_similar_to_vllm_granularity0 SKIPPED 2025-09-09T15:46:22.5208573Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_and_copy_similar_to_vllm_granularity1 SKIPPED 2025-09-09T15:46:22.5210038Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_granularity0 SKIPPED 2025-09-09T15:46:22.5211390Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_granularity1 SKIPPED 2025-09-09T15:46:22.5212839Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_preserves_aliasing_granularity0 SKIPPED 2025-09-09T15:46:22.5214385Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_slice_preserves_aliasing_granularity1 SKIPPED 2025-09-09T15:46:22.5215879Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity0_sizes0 SKIPPED 2025-09-09T15:46:22.5217334Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity0_sizes1 SKIPPED 2025-09-09T15:46:22.5218791Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity0_sizes2 SKIPPED 2025-09-09T15:46:22.5220233Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity1_sizes0 SKIPPED 2025-09-09T15:46:22.5221683Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity1_sizes1 SKIPPED 2025-09-09T15:46:22.5223136Z test/quantization/quantize_/workflows/float8/test_float8_tensor.py::TestFloat8Tensor::test_to_device_granularity1_sizes2 SKIPPED 2025-09-09T15:46:22.5224800Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_linear_config0_sizes0 PASSED 2025-09-09T15:46:22.5226380Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_linear_config0_sizes1 PASSED 2025-09-09T15:46:22.5228022Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_linear_config0_sizes2 PASSED 2025-09-09T15:46:22.5229576Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_module_path_config0 PASSED 2025-09-09T15:46:22.5231122Z test/quantization/quantize_/workflows/int4/test_int4_marlin_sparse_tensor.py::TestInt4MarlinSparseTensor::test_to_device_config0 SKIPPED 2025-09-09T15:46:22.5232670Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_bfloat16_group_size_128 PASSED 2025-09-09T15:46:22.5234244Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_bfloat16_group_size_32 PASSED 2025-09-09T15:46:22.5235809Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_bfloat16_group_size_64 PASSED 2025-09-09T15:46:22.5237462Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float16_group_size_128 PASSED 2025-09-09T15:46:22.5239037Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float16_group_size_32 PASSED 2025-09-09T15:46:22.5240605Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float16_group_size_64 PASSED 2025-09-09T15:46:22.5242178Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float32_group_size_128 PASSED 2025-09-09T15:46:22.5243789Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float32_group_size_32 PASSED 2025-09-09T15:46:22.5245378Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes0_float32_group_size_64 PASSED 2025-09-09T15:46:22.5246968Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_bfloat16_group_size_128 PASSED 2025-09-09T15:46:22.5248551Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_bfloat16_group_size_32 PASSED 2025-09-09T15:46:22.5250124Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_bfloat16_group_size_64 PASSED 2025-09-09T15:46:22.5251709Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float16_group_size_128 PASSED 2025-09-09T15:46:22.5253284Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float16_group_size_32 PASSED 2025-09-09T15:46:22.5254847Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float16_group_size_64 PASSED 2025-09-09T15:46:22.5256425Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float32_group_size_128 PASSED 2025-09-09T15:46:22.5257986Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float32_group_size_32 PASSED 2025-09-09T15:46:22.5259552Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes1_float32_group_size_64 PASSED 2025-09-09T15:46:22.5261331Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_bfloat16_group_size_128 PASSED 2025-09-09T15:46:22.5262971Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_bfloat16_group_size_32 PASSED 2025-09-09T15:46:22.5264767Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_bfloat16_group_size_64 PASSED 2025-09-09T15:46:22.5266330Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float16_group_size_128 PASSED 2025-09-09T15:46:22.5267887Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float16_group_size_32 PASSED 2025-09-09T15:46:22.7297456Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float16_group_size_64 PASSED 2025-09-09T15:46:22.7299963Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float32_group_size_128 PASSED 2025-09-09T15:46:22.7302407Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float32_group_size_32 PASSED 2025-09-09T15:46:22.7304323Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_linear_sizes2_float32_group_size_64 PASSED 2025-09-09T15:46:22.7305491Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_module_path_bfloat16 PASSED 2025-09-09T15:46:22.7306605Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_module_path_float16 PASSED 2025-09-09T15:46:22.7307714Z test/quantization/quantize_/workflows/int4/test_int4_opaque_tensor.py::TestInt4OpaqueTensor::test_module_path_float32 PASSED 2025-09-09T15:46:22.7308923Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_bfloat16_group_size_128 SKIPPED 2025-09-09T15:46:22.7310211Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_bfloat16_group_size_32 SKIPPED 2025-09-09T15:46:22.7311471Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_bfloat16_group_size_64 SKIPPED 2025-09-09T15:46:22.7312746Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_float16_group_size_128 SKIPPED 2025-09-09T15:46:22.7314027Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_float16_group_size_32 SKIPPED 2025-09-09T15:46:22.7315286Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes0_float16_group_size_64 SKIPPED 2025-09-09T15:46:22.7316644Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_bfloat16_group_size_128 SKIPPED 2025-09-09T15:46:22.7317927Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_bfloat16_group_size_32 SKIPPED 2025-09-09T15:46:22.7319185Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_bfloat16_group_size_64 SKIPPED 2025-09-09T15:46:22.7320480Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_float16_group_size_128 SKIPPED 2025-09-09T15:46:22.7321734Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_float16_group_size_32 SKIPPED 2025-09-09T15:46:22.7323230Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes1_float16_group_size_64 SKIPPED 2025-09-09T15:46:22.7324610Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_bfloat16_group_size_128 SKIPPED 2025-09-09T15:46:22.7325876Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_bfloat16_group_size_32 SKIPPED 2025-09-09T15:46:22.7327137Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_bfloat16_group_size_64 SKIPPED 2025-09-09T15:46:22.7328392Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_float16_group_size_128 SKIPPED 2025-09-09T15:46:22.7329653Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_float16_group_size_32 SKIPPED 2025-09-09T15:46:22.7330910Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_linear_sizes2_float16_group_size_64 SKIPPED 2025-09-09T15:46:22.7332104Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_module_path_bfloat16 SKIPPED 2025-09-09T15:46:22.7333242Z test/quantization/quantize_/workflows/int4/test_int4_plain_int32_tensor.py::Int4PlainInt32Tensor::test_module_path_float16 SKIPPED 2025-09-09T15:46:22.7334404Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_bmm_bmm_config0 SKIPPED 2025-09-09T15:46:22.7335568Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_bmm_bmm_config1 SKIPPED 2025-09-09T15:46:22.7336740Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_linear_config0 SKIPPED 2025-09-09T15:46:22.7337893Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_linear_config1 SKIPPED 2025-09-09T15:46:22.7339079Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_module_path_config0 SKIPPED 2025-09-09T15:46:22.7340281Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_module_path_config1 SKIPPED 2025-09-09T15:46:22.7341477Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_to_device_config0 SKIPPED 2025-09-09T15:46:22.7342665Z test/quantization/quantize_/workflows/int4/test_int4_preshuffled_tensor.py::TestInt4PreshuffledTensor::test_to_device_config1 SKIPPED 2025-09-09T15:46:22.7343765Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_activation_prescaling SKIPPED 2025-09-09T15:46:22.7344750Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_bmm SKIPPED 2025-09-09T15:46:22.7345647Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_cat_sizes0 SKIPPED 2025-09-09T15:46:22.7346555Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_cat_sizes1 SKIPPED 2025-09-09T15:46:22.7347467Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_cat_sizes2 SKIPPED 2025-09-09T15:46:22.7348364Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_linear SKIPPED 2025-09-09T15:46:22.7349312Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_moe_weight_reshape_ops SKIPPED 2025-09-09T15:46:22.7350348Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_slice SKIPPED 2025-09-09T15:46:22.7351322Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_slice_and_copy_similar_to_vllm SKIPPED 2025-09-09T15:46:22.7352514Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_slice_preserves_aliasing SKIPPED 2025-09-09T15:46:22.7353514Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_to_device_sizes0 SKIPPED 2025-09-09T15:46:22.7354516Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_to_device_sizes1 SKIPPED 2025-09-09T15:46:22.7355472Z test/quantization/quantize_/workflows/int4/test_int4_tensor.py::TestInt4Tensor::test_to_device_sizes2 SKIPPED 2025-09-09T15:46:22.7356674Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_cant_initialize_in_cpu SKIPPED 2025-09-09T15:46:22.7358012Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_different_group_sizes_group_size_128 SKIPPED 2025-09-09T15:46:22.7359387Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_different_group_sizes_group_size_32 SKIPPED 2025-09-09T15:46:22.7360740Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_different_group_sizes_group_size_64 SKIPPED 2025-09-09T15:46:22.7362028Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_error_conditions SKIPPED 2025-09-09T15:46:22.7363272Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes0_config0 SKIPPED 2025-09-09T15:46:22.7364753Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes0_config1 SKIPPED 2025-09-09T15:46:22.7366020Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes1_config0 SKIPPED 2025-09-09T15:46:22.7520301Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes1_config1 SKIPPED 2025-09-09T15:46:22.7521576Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes2_config0 SKIPPED 2025-09-09T15:46:22.7522868Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_linear_sizes2_config1 SKIPPED 2025-09-09T15:46:22.7524182Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_mm_int4wo_device_cuda_bfloat16 SKIPPED 2025-09-09T15:46:22.7525471Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_module_path_config0 SKIPPED 2025-09-09T15:46:22.7526736Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_module_path_config1 SKIPPED 2025-09-09T15:46:22.7528071Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_and_copy_similar_to_vllm_config0 SKIPPED 2025-09-09T15:46:22.7529457Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_and_copy_similar_to_vllm_config1 SKIPPED 2025-09-09T15:46:22.7530752Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_config0 SKIPPED 2025-09-09T15:46:22.7532114Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_config1 SKIPPED 2025-09-09T15:46:22.7533501Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_preserves_aliasing_config0 SKIPPED 2025-09-09T15:46:22.7534855Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_slice_preserves_aliasing_config1 SKIPPED 2025-09-09T15:46:22.7536104Z test/quantization/quantize_/workflows/int4/test_int4_tile_packed_to_4d_tensor.py::TestInt4TilePackedTo4dTensor::test_to_device SKIPPED 2025-09-09T15:46:22.7538146Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7546786Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7549553Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7552317Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7555107Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7557942Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7560704Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7563465Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7566515Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7569613Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7572421Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7575204Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7578082Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7580843Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7583670Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7586537Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7682297Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7685104Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7687871Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7691437Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7694265Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7697154Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7700070Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7702846Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7705621Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7708382Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7711207Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7714090Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7716973Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7719892Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7722654Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7725418Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7728285Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7731164Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7733977Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7736748Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7739524Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7843660Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7846505Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7849605Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7852421Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7855199Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7858034Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7860808Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7863630Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7866650Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7869486Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7872301Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7875154Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7878312Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7881152Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7883945Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7886791Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7889576Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7892370Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7895156Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.7897944Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.7900739Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8008391Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8011482Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8014420Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8017234Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8020123Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8022933Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8025798Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8028723Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8031571Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8034402Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8037263Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8040078Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8043080Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8045994Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8048868Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8051754Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8054574Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8057396Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8060261Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8063192Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8066214Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8168931Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8172002Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8174813Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8177701Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8180697Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8183574Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8186399Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8189207Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8192028Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8194908Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8197917Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8200905Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8203720Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8206548Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8209408Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8212288Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8215214Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8218087Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8220959Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8223815Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8226670Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8328733Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8334211Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8337067Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8339999Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8342911Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8345886Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8348789Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8351656Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8354516Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8357442Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8360507Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8363470Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8366582Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8369551Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8372408Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8375281Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8378186Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8381153Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8384094Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8386973Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8390004Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8486985Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8489947Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8493004Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8495903Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8498788Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8501644Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8504511Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8507460Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8510482Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8522815Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8526436Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8529302Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8532211Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8535119Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8538096Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8540990Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8543871Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8546726Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8549575Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8552615Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8555634Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.bfloat16, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8654528Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8657345Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8660119Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8662902Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8665832Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8668624Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8671401Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8674185Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8677708Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8680602Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8683445Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8686301Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8689080Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8691858Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8694691Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8697586Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8700422Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8703203Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8706160Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8708941Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8711784Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8714798Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8822622Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8827022Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8832578Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8836109Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8839044Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8841945Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8844987Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8847789Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8850573Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8853453Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8856307Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8859209Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8862068Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8865079Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8867892Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8870687Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8873713Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8876718Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8879558Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8882434Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8885227Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8888013Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8982985Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8985919Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_auto', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8988780Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8991613Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.8994680Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.8997703Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_kleidiai', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9000576Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9003487Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9006297Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9009097Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9011911Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9014728Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9017544Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9020351Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9023205Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9026328Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9029186Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9032021Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9034938Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9037776Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9040650Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9143986Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9146964Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9149753Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9152617Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9155677Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9158623Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9161501Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9164618Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9167405Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9170183Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9172947Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9175758Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9178638Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9181448Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9184411Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9187184Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9189949Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9192827Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9195755Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9198675Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9201583Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9305404Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9308228Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9311045Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9314197Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9317157Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9319938Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9322770Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9325589Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9328413Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9331300Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': 'torchao_lowbit', 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9334150Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9336977Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9339792Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9342744Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9345569Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9348398Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9351258Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9354097Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9357103Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9360030Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9362898Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9465699Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9468566Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9471632Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9474514Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9477517Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9480467Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9483292Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9486113Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9488925Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9491785Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9494760Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9497618Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9500576Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9503392Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9506206Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9509140Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9512068Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9514940Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9517844Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:22.9520659Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:22.9523460Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1573165Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1577437Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1581208Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1584916Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1588637Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1592211Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1596022Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1599925Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1603659Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1607216Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1610770Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1614473Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1618115Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:23.1621809Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_accuracy_{'model_dtype': torch.float32, 'packing_format': , 'compute_target': None, 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:23.1624361Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_export_compile_aoti SKIPPED 2025-09-09T15:46:23.1625772Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_moe_quant_intx SKIPPED 2025-09-09T15:46:23.1627594Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_serialization_{'packing_format': , 'compute_target': 'aten'} SKIPPED 2025-09-09T15:46:23.1629835Z test/quantization/quantize_/workflows/intx/test_intx_opaque_tensor.py::TestIntxOpaqueTensor::test_serialization_{'packing_format': , 'compute_target': 'torchao_auto'} SKIPPED 2025-09-09T15:46:23.1631701Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_embedding PASSED 2025-09-09T15:46:23.1633362Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_export_int8_dyn_act_intx_weight_config PASSED 2025-09-09T15:46:23.1635254Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_export_int8_dyn_act_intx_weight_config_with_unwrap PASSED 2025-09-09T15:46:23.1637102Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_export_intx_weight_only_config PASSED 2025-09-09T15:46:23.1639811Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.1643321Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.1646832Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2553402Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2557020Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2560600Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2564383Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2567890Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2571433Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2574994Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2578510Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2582034Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2585521Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2589258Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2592794Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2596428Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2600015Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2603493Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2607073Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2610564Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2614057Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2617594Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.2621100Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.2624583Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.2628262Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3550808Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3554380Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3558143Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3561708Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3565417Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3568961Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3572477Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3576036Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3579544Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3583335Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3586869Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3590303Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3593793Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3597379Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3600893Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3604413Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3607908Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3611417Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3614917Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.3618556Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.3622070Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.3625619Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4543400Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4546964Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4550476Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4554041Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4557646Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4561173Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4564921Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4568564Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4572402Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4575929Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4579531Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4583052Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4586555Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4590007Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4593475Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4597139Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4600654Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4604176Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4607875Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.4611392Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.4614893Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.4618558Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5511299Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5514866Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5518473Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5522039Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5525581Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5529131Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5532910Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5536437Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5539947Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5543571Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5547161Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5550683Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5554227Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5557887Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5561428Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5565205Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5568990Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5572539Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5576148Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.5580091Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.5583797Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.5587479Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6506798Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6510302Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6513774Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6517397Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6520907Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6524723Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6528243Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6531766Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6535368Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6538896Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6542418Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6545935Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6549457Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6552960Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6556619Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6560359Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6564010Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6567514Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6571084Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.6574648Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.6578344Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.6581837Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7494651Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7498246Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7501794Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7505574Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7509224Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7512772Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7516456Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7519956Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7523498Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7527054Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7530590Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7534127Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7537688Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7541230Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7544965Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7548476Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7552030Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7555673Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7559312Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.7562856Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.7566758Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.7570291Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8471374Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8478307Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8482085Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8485666Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8489253Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8492853Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8496365Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8499905Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8503418Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8506953Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8510473Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8513943Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8517761Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8521292Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8524789Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8528372Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8531853Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8535361Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8538879Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.8542436Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.8546062Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.8549566Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9451608Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9455192Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9458707Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9462325Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9466216Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9469768Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9473354Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9477012Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9480544Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9484084Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9487733Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9491355Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9494969Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9498498Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9502097Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9505644Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9509288Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9512805Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9516339Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:23.9519875Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:23.9523375Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:23.9527051Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0409829Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0413401Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0417066Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0420653Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0424198Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0427725Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0431234Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0434802Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0438444Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0442193Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0445797Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0449337Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0452950Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0456492Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0460045Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0463569Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0467326Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0470888Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0474433Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.0478194Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.0481944Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.0485525Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.1941903Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.1945584Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.1949072Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.1952577Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.1956125Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.1959722Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.1963211Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.1966874Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.1970525Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.1974019Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.1977511Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.1981098Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.1984680Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.1988249Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.1991721Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_intx_unpacked_v2_is_close_to_qdq_v1_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.1994099Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_linear PASSED 2025-09-09T15:46:24.1996548Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.1999645Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.2002728Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.2005927Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.2008933Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.2012011Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.2015203Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4689404Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4692566Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4695673Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4698756Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4701810Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4704894Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4707952Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4711030Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4714404Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4717579Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4720659Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4723839Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4726939Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4730007Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4733113Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4736192Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4739269Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4742342Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4745423Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4748543Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4751750Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4754829Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.4758052Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.4761211Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.4764503Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7421207Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7424353Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7427515Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7430615Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7433690Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7436878Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7440204Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7443225Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7446259Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7449251Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7452293Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7455413Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7458489Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7461571Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7464862Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7467979Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7471051Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7474131Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7477528Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7480628Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7483698Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7486850Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:24.7489984Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:24.7493084Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:24.7496206Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0115815Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0119037Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0122149Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0125230Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0128491Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0131708Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0134817Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0137962Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0141030Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0144045Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0147044Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0150102Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0153161Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0156332Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0159512Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0162582Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0166808Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0169945Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0173033Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0176120Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0179255Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0182230Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.0185273Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.0188310Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.0191403Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2829864Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2833009Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2836103Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2839600Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2842659Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2845681Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2848705Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2851788Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2854797Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2857843Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2860911Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2864179Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2867270Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2870348Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2873444Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2876892Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2879972Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2883058Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2886349Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2889425Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2892525Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2895583Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.2898672Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.2901748Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.2904834Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5626982Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5630274Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5633515Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5636712Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5639808Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5644985Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5648089Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5651188Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5654270Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5657413Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5660521Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5663626Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5666908Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5670141Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5673316Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5676573Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5679628Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5682774Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5685867Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5688950Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5692028Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5695082Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.5698154Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.5701215Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.5704274Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8367189Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8370336Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8373405Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8376519Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8379700Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8382800Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8385895Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8388989Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8392099Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8395213Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8398377Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8401454Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8404719Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8407855Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8411070Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8414194Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8417278Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8420343Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8423419Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8426482Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8429559Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8432660Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:25.8435752Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:25.8439060Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:25.8442273Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1102844Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1105973Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1109164Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1112183Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1115180Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1118332Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1121564Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1124758Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1128220Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1131443Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1134778Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1137982Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1141267Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1144492Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1158208Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1161521Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1164791Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1167868Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1170925Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1173985Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1177056Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1180112Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.1183426Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.1186599Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.1189785Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3821103Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3824432Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3827507Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3830593Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3833675Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3836793Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3839880Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3842975Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3846088Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3849444Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3852553Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3855660Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3858814Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3861913Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3865300Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3868421Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3871539Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3874636Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3877786Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3880873Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3884017Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3888140Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.3891125Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.3894164Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.3897299Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.5282048Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.5284495Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.5286871Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.5289251Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.5291626Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.5294007Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.5296382Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} PASSED 2025-09-09T15:46:26.5299043Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} PASSED 2025-09-09T15:46:26.5301412Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_qat_int8_dyn_act_intx_weight_config_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} PASSED 2025-09-09T15:46:26.5303301Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_serialization_int8_dyn_act_intx_weight_config PASSED 2025-09-09T15:46:26.5304736Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_serialization_intx_weight_only_config PASSED 2025-09-09T15:46:26.5306017Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_slice PASSED 2025-09-09T15:46:26.5307276Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_slice_and_copy_ PASSED 2025-09-09T15:46:26.5308483Z test/quantization/quantize_/workflows/intx/test_intx_unpacked_to_int8_tensor.py::TestIntxUnpackedToInt8Tensor::test_to_dtype PASSED 2025-09-09T15:46:26.5309531Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_2_bias_False SKIPPED 2025-09-09T15:46:26.5310438Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_2_bias_True SKIPPED 2025-09-09T15:46:26.5311347Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_3_bias_False SKIPPED 2025-09-09T15:46:26.5312254Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_concat_linear_cpu_x_dim_3_bias_True SKIPPED 2025-09-09T15:46:26.5313228Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5314259Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5315280Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5316386Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5317876Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5319139Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5320162Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5321168Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_2_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5322168Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5323187Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5324195Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5325198Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5326302Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5327360Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5328365Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5329353Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_bfloat16_x_dim_3_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5330357Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5331371Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5332371Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5333416Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5334422Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5335416Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5336421Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5337401Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_2_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5338413Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5339429Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5340431Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5341423Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:26.5342424Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:26.5343424Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0500976Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0502013Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float16_x_dim_3_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0503035Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0504062Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0505067Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0506056Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0507045Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0508213Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0509392Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0510384Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_2_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0511366Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0512359Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0513345Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0514332Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_False_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0515416Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_160_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0516486Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_160_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0517475Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_1_sym_quant_a_False SKIPPED 2025-09-09T15:46:54.0518447Z test/quantization/test_da8w4_cpu.py::TestDa8w4Cpu::test_8da4w_cpu_float32_x_dim_3_bias_True_bs_1_sym_quant_a_True SKIPPED 2025-09-09T15:46:54.0519304Z test/quantization/test_gptq.py::TestGPTQ::test_gptq_quantizer_int4_weight_only SKIPPED 2025-09-09T15:46:54.0520077Z test/quantization/test_gptq.py::TestMultiTensorFlow::test_multitensor_add_tensors PASSED 2025-09-09T15:46:54.0520902Z test/quantization/test_gptq.py::TestMultiTensorFlow::test_multitensor_inplace_operation PASSED 2025-09-09T15:46:54.0521716Z test/quantization/test_gptq.py::TestMultiTensorFlow::test_multitensor_pad_unpad PASSED 2025-09-09T15:46:54.0522525Z test/quantization/test_gptq.py::TestMultiTensorInputRecorder::test_gptq_with_input_recorder layers.0.attention.wqkv.weight 2025-09-09T15:46:54.0523141Z layers.0.attention.wo.weight 2025-09-09T15:46:54.0523410Z layers.0.feed_forward.w1.weight 2025-09-09T15:46:54.0523677Z layers.0.feed_forward.w3.weight 2025-09-09T15:46:54.0523940Z layers.0.feed_forward.w2.weight 2025-09-09T15:46:54.0524198Z layers.1.attention.wqkv.weight 2025-09-09T15:46:54.0524465Z layers.1.attention.wo.weight 2025-09-09T15:46:54.0524720Z layers.1.feed_forward.w1.weight 2025-09-09T15:46:54.0524981Z layers.1.feed_forward.w3.weight 2025-09-09T15:46:54.0525231Z layers.1.feed_forward.w2.weight 2025-09-09T15:46:54.0525479Z output.weight 2025-09-09T15:46:54.0525713Z PASSED 2025-09-09T15:46:54.0526301Z test/quantization/test_gptq.py::TestMultiTensorInputRecorder::test_multitensor_input_recorder PASSED 2025-09-09T15:46:54.0527385Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_aten SKIPPED 2025-09-09T15:46:54.0528601Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_kleidiai SKIPPED 2025-09-09T15:46:54.0530921Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0534285Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0537563Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0540801Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0544161Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0547538Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0550843Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0554107Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0557449Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0560696Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0631742Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0635273Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0638650Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0641986Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0645233Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0648496Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0651799Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0655165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0658464Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0661763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0665385Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0668646Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0672027Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0675388Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0678773Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0682035Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0685295Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0688547Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0691928Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0761050Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0764575Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0767938Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0771194Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0774450Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0777766Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0781142Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0784450Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0787811Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0791438Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0795567Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0799151Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0802521Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0805838Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0809103Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0812353Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0815614Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0818918Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0822423Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0891702Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0895047Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0898451Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0901764Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0905133Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0908550Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int1, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0911918Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0915310Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0918842Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0922146Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0925518Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0928983Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int2, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0932357Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0935680Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0938993Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0942304Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0945716Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.0949214Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int3, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.0952580Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1020487Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1023816Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1027136Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1030503Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1033948Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int4, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1037365Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1040783Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1044207Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1047529Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1050971Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1054406Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int5, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1057779Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1061111Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1064663Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1068051Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1071497Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1075959Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int6, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1079380Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1082778Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1187496Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1190824Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1194188Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1197703Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int7, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1201081Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1204519Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1207944Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1211255Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1215655Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerAxis(axis=0)} SKIPPED 2025-09-09T15:46:54.1220022Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_accuracy_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.UNIVERSAL), 'weight_dtype': torch.int8, 'weight_mapping_type': , 'weight_granularity': PerGroup(group_size=128)} SKIPPED 2025-09-09T15:46:54.1222883Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_export_QDQLayout SKIPPED 2025-09-09T15:46:54.1224784Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_export_compile_aoti_PackedLinearInt8DynamicActivationIntxWeightLayout SKIPPED 2025-09-09T15:46:54.1226965Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_export_dynamic_shape_PackedLinearInt8DynamicActivationIntxWeightLayout SKIPPED 2025-09-09T15:46:54.1229396Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1231480Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1233557Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1235680Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1237783Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1239981Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1242043Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1244095Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1246195Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 128, 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1248301Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1250358Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1252407Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1359287Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1361371Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1363419Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1365676Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1367737Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1369776Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 32, 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1371839Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1374056Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1376199Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1378264Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1380305Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1382420Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1384490Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1386544Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1388607Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynActInt4WeightQATQuantizer_{'group_size': 64, 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1390830Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynamicActivationInt4WeightConfig_{'group_size': 32, 'mapping_type': , 'act_mapping_type': } SKIPPED 2025-09-09T15:46:54.1393186Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_Int8DynamicActivationInt4WeightConfig_{'group_size': 64, 'mapping_type': , 'act_mapping_type': } SKIPPED 2025-09-09T15:46:54.1395871Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1398919Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1401883Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1404983Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1407948Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1410893Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1413925Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1416883Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1419836Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1505558Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1508955Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1511903Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1514864Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1518086Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1521031Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1523981Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1526984Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1529920Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1532883Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1535894Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1538844Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1541796Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1544749Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1547866Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1550817Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1553764Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1556806Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1559754Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1562714Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1565869Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1645619Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1648575Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1651496Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1654645Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1657590Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1660529Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1663550Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1666778Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1669739Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1672707Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1675667Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1678681Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1681625Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1684801Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1687761Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1690718Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1693739Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1696679Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1699649Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1702591Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1705542Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1789615Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1795114Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1798327Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1801289Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1804267Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1807300Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1810256Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1813217Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1816169Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1819127Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1822078Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1825073Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1828149Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1831100Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1834056Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1837145Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1840096Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1843030Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1845970Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1848910Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1851842Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1927954Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1931134Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1934101Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1937090Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1940107Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1943075Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1946051Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1949009Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1951975Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1955024Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1958176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1973548Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1976585Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1979503Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1982474Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1985364Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1988271Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.1991165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.1994080Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.1997243Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2000939Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2070078Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2073012Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2075933Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2078976Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2081893Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2084817Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2087750Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2090681Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2093597Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2096580Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2099601Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2102521Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2105449Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2108415Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2111319Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int1, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2114266Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2117276Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2120241Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2123206Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2126201Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2129232Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2210359Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2213332Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2216400Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2219336Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2222290Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2225280Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2228224Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2231163Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2234159Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2237326Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2240261Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2243196Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2246200Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2249143Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2252090Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2255030Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2257966Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2260893Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2264071Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2267120Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2270058Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2354102Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2357537Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2360471Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2363407Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2366478Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2369400Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2372329Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2375331Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2378359Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2381298Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2384252Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2387267Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2390209Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2393151Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2396130Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2399139Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2402080Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2405105Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2408160Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2411098Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2414035Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2492522Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2495475Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2498406Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2501334Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2504274Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2507192Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2510219Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2513286Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2516217Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2519306Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2522230Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2525156Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2528090Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2531005Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2533934Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2536872Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2539853Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2542854Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2545770Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2548735Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2551647Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2637060Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2640025Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2642945Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2645886Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2648839Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2651884Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2654963Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2657910Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2660906Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2664068Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2667067Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2670017Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2672968Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2675964Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2678986Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2682121Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2685116Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2688046Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2691044Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2693972Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2696918Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2779352Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2782316Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2785272Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2788218Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2791370Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2794315Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2797326Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2800355Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2803273Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2806209Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2809139Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2812060Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2814990Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2817901Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2820936Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2823859Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2826776Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2829733Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int2, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2832673Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2835637Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2838640Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2919024Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2922024Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2924984Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2929030Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2932008Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2934957Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2937988Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2940943Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2943893Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2946857Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2949809Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2952764Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2955726Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2958926Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2961870Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2964974Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2968021Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2970967Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.2973915Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.2976854Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.2979788Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3060335Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3063295Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3066650Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3069602Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3072544Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3075614Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3078604Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3081553Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3084470Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3087403Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3090337Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3093263Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3096352Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3099304Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3102262Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3105250Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3108191Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3111134Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3114085Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3117081Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3120031Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3202149Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3205339Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3208287Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3211241Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3214263Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3217205Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3220154Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3223090Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3226029Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3228982Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3231934Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3235005Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3238017Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3240964Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3243942Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3246878Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3249821Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3252745Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3255683Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3258621Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3261564Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3343691Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3347080Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3350002Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3352983Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3355901Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3358882Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3361807Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3364899Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3367848Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3370792Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3373913Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3376850Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3379787Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3382809Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3385797Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3388740Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3391684Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3394612Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3397654Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3400588Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3403676Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3483096Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3486049Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3489087Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3492026Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3494982Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3497924Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3500860Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3503796Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3506721Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3509825Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3512773Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3515696Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3518745Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3521684Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3524626Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3527563Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3530483Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3533410Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3536342Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3539406Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3542342Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int3, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3623584Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3627228Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3630191Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3633165Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3636121Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3639143Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3642096Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3645049Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3648177Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3651142Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3654103Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3657090Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3660027Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3662978Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3666140Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3669089Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3672022Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3674960Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3678176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3681135Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3684084Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3765257Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3768217Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3771167Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3774105Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3777059Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3779996Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3782931Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3786086Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3789041Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3791986Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3794996Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3797995Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3800945Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3803872Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3806800Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3809750Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3812692Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3815777Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3818732Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3821691Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3824666Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3904587Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3907550Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3910909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3914624Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3917992Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3920928Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3924064Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3926995Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3929928Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3932913Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3935838Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3938766Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3941700Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3944643Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3947580Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3950587Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3953590Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3956579Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.3959525Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.3962517Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.3965716Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4044328Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4047311Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4050277Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4053229Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4056281Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4059330Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4062269Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4065359Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4068371Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4071337Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4074323Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4077330Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4080294Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4083251Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4086273Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4089407Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4092352Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4095304Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4098306Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4101263Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4104222Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4185251Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4188232Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4191193Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4194238Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4197395Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4200344Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4203291Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4206340Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4209294Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4212264Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4215212Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4218157Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4221097Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4224093Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4227226Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4230183Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4233117Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4236155Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4239152Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4242096Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4245029Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4324491Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4327500Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4330537Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int4, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4341567Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4344556Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4347522Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4350536Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4353481Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4356535Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4359475Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4362413Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4365989Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4369803Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4372857Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4375795Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4378735Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4381720Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4384648Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4387576Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4390499Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4393423Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4466089Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4469146Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4472202Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4475189Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4478272Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4481195Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4484128Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4487067Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4489987Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4492925Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4495855Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4498829Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4501830Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4504744Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4507788Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4510705Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4513621Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4516645Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4519579Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4522540Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4525486Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4606528Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4609622Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4612570Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4615579Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4618527Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4621471Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4624431Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4627379Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4630331Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4633279Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4637022Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4639971Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4642919Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4645909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4648854Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4651810Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4654767Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4657711Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4660664Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4663598Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4666883Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4747016Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4749992Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4753054Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4756002Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4759044Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4761991Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4765139Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4768088Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4771007Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4774153Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4777092Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4780013Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4783034Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4785985Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4788957Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4791917Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4794864Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4797888Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4800837Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4803918Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4806862Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4888176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4891279Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4894236Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4897184Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4900133Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4903069Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4906030Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4908986Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4912098Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4915038Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4918041Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4921048Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4923999Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4926931Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4929868Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4932798Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4935742Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4938679Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.4941740Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.4944684Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.4947615Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5030573Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5035689Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5038669Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5041595Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5044514Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5047437Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int5, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5050386Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5053588Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5057323Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5061052Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5064220Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5067232Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5070190Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5073143Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5076145Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5079160Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5082121Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5085264Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5088233Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5091176Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5094194Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5170200Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5175428Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5178365Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5181328Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5184286Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5187236Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5190436Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5193381Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5196426Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5199437Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5202378Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5205315Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5208249Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5211198Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5214154Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5217140Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5220185Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5223111Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5226103Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5229069Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5231983Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5308614Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5311573Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5314530Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5317613Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5320551Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5323723Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5326675Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5329604Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5332597Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5335537Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5338482Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5341423Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5344355Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5347294Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5350216Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5353274Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5356208Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5359186Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5362167Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5365276Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5368219Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5447702Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5450641Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5453584Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5456515Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5459654Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5462596Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5465679Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5468690Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5471629Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5474560Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5477531Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5480439Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5483360Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5486282Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5489360Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5492304Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5495250Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5498277Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5501226Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5504177Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5507163Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5590265Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5595465Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5598462Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5601615Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5604562Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5607505Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5610504Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5613433Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5616373Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5619310Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5622238Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5625167Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5628103Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5631188Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5634145Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5637143Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5640117Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5643044Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5645979Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5648905Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5651828Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5730582Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5733554Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5736703Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5739650Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5742576Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5745561Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5748488Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5751422Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5754347Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int6, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5757361Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5760324Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5763274Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5767276Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5770232Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5773179Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5776201Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5779141Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5782105Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5785056Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5788019Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5790967Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5869910Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5873134Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5876136Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5879152Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5882183Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5885132Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5888101Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5891065Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5894022Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5896986Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5899937Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5903032Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5906042Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5908993Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5911980Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5914944Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5918000Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5920949Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.5923901Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.5926845Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.5929834Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6010458Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6013409Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6016429Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6019448Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6022421Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6025392Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6028351Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6031321Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6034262Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6037389Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6040466Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6043419Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6046427Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6049424Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6052377Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6055333Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6058284Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6061227Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6064320Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6067402Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6070459Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6149682Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6153064Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6156106Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6159112Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6162052Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6165196Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6168144Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6171093Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6174114Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6177175Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6180119Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6183063Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6186065Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6188992Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6191926Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6194863Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6197931Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6200855Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6203909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6207013Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6209980Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6290983Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6294478Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6298237Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6301970Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6305689Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6309416Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6313122Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6316880Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6319962Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6322937Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6325903Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6329589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6332534Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6342951Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6345891Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6348829Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6351755Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6354750Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6357820Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6360735Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6363653Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6431222Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6434141Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6437109Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6440026Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6442943Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6445860Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6448876Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6451912Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6454832Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6457757Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6460742Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6463651Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int7, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6466792Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6469749Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6472699Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6475693Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6478756Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6481803Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6484751Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6487826Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6490756Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6570800Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6573784Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6576718Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6579667Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6582589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6585610Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6588682Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6591613Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6594613Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6597617Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6600575Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6603514Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6606449Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6609383Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6612305Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6615371Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6618315Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6621243Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6624233Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6627178Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6630113Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6712697Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6716859Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6719790Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6722727Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6725863Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6728787Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 128, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6731734Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6734769Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6737713Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6740664Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6743603Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6746592Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6749537Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6752475Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6755592Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6758594Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6761554Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6764686Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6767681Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6770625Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6773545Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6852915Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6855877Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6858825Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6862011Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6865115Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6868055Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6871083Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6874023Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6877009Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6879941Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6882865Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6885813Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6888746Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6891853Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6894786Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6897708Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6900668Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6903589Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6906507Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6909433Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6912336Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 32, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.6991763Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.6994718Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.6997960Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.7000923Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.7003880Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.7006873Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.7009816Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.7012772Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.7015707Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.7018651Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.7021596Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.7024528Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.7027590Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.7030529Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.7033462Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.7036533Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.7039457Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.7042383Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T15:46:54.7045375Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T15:46:54.7048327Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T15:46:54.7051267Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T16:19:30.2343880Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T16:19:30.2348389Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T16:19:30.2351370Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T16:19:30.2354291Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T16:19:30.2357959Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T16:19:30.2361062Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T16:19:30.2364189Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T16:19:30.2367128Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T16:19:30.2370050Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.bfloat16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T16:19:30.2372970Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T16:19:30.2375909Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float16} SKIPPED 2025-09-09T16:19:30.2379026Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float16, 'model_dtype': torch.float32} SKIPPED 2025-09-09T16:19:30.2381950Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.bfloat16} SKIPPED 2025-09-09T16:19:30.2384869Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float16} SKIPPED 2025-09-09T16:19:30.2387895Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_identical_to_IntXQuantizationAwareTrainingConfig_{'weight_dtype': torch.int8, 'group_size': 64, 'mapping_type': , 'act_mapping_type': , 'scale_dtype': torch.float32, 'model_dtype': torch.float32} SKIPPED 2025-09-09T16:19:30.2389906Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_moe_quant_intx SKIPPED 2025-09-09T16:19:30.2391797Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_serialization_{'layout': PackedLinearInt8DynamicActivationIntxWeightLayout(group_size=None, bit_width=None, has_weight_zeros=None, has_bias=None, target=Target.AUTO)} SKIPPED 2025-09-09T16:19:30.2393679Z test/quantization/test_int8_dynamic_activation_intx_weight_config_v1.py::TestInt8DynamicActivationIntxWeight::test_serialization_{'layout': QDQLayout()} SKIPPED 2025-09-09T16:19:30.2394802Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_base_0_single_token SKIPPED 2025-09-09T16:19:30.2395720Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_base_1_multiple_tokens SKIPPED 2025-09-09T16:19:30.2396865Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_fake_dim_0_single_token SKIPPED 2025-09-09T16:19:30.2398127Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8dq_fake_dim_1_multiple_tokens SKIPPED 2025-09-09T16:19:30.2399005Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_base_0_single_token SKIPPED 2025-09-09T16:19:30.2399865Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_base_1_multiple_tokens SKIPPED 2025-09-09T16:19:30.2400748Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_fake_dim_0_single_token SKIPPED 2025-09-09T16:19:30.2401632Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_fp8wo_fake_dim_1_multiple_tokens SKIPPED 2025-09-09T16:19:30.2402504Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_base_0_single_token SKIPPED 2025-09-09T16:19:30.2403370Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_base_1_multiple_tokens SKIPPED 2025-09-09T16:19:30.2404243Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_fake_dim_0_single_token PASSED 2025-09-09T16:19:30.2405129Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int4wo_fake_dim_1_multiple_tokens PASSED 2025-09-09T16:19:30.2406066Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8dq_base_0_multiple_tokens PASSED 2025-09-09T16:19:30.2407081Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8dq_fake_dim_0_multiple_tokens PASSED 2025-09-09T16:19:30.2407996Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_0_single_token PASSED 2025-09-09T16:19:30.2408848Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_1_multiple_tokens PASSED 2025-09-09T16:19:34.1776261Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_cpu_0_single_token cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1777337Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 33, in forward 2025-09-09T16:19:34.1778042Z scores = self.router(x) # [T, E] 2025-09-09T16:19:34.1778240Z 2025-09-09T16:19:34.1778394Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1779106Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 34, in forward 2025-09-09T16:19:34.1780877Z scores = F.softmax(scores, dim=-1) 2025-09-09T16:19:34.1781068Z 2025-09-09T16:19:34.1781229Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1781919Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 34, in forward 2025-09-09T16:19:34.1782571Z scores = F.softmax(scores, dim=-1) 2025-09-09T16:19:34.1782763Z 2025-09-09T16:19:34.1782870Z cudagraph partition due to non gpu ops 2025-09-09T16:19:34.1783214Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1783895Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 35, in forward 2025-09-09T16:19:34.1784545Z scores, expert_indices = torch.topk( 2025-09-09T16:19:34.1784743Z 2025-09-09T16:19:34.1784894Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1785574Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 35, in forward 2025-09-09T16:19:34.1786230Z scores, expert_indices = torch.topk( 2025-09-09T16:19:34.1786427Z 2025-09-09T16:19:34.1786569Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1787271Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 38, in forward 2025-09-09T16:19:34.1787990Z scores /= scores.sum(dim=-1, keepdim=True).to(x.dtype) # [T, A] 2025-09-09T16:19:34.1788260Z 2025-09-09T16:19:34.1788408Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1789089Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1789794Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1790490Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 99, in forward 2025-09-09T16:19:34.1791139Z y1 = F.silu(F.linear(x, w1[index])) 2025-09-09T16:19:34.1791329Z 2025-09-09T16:19:34.1791481Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1792160Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1792862Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1793558Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 100, in forward 2025-09-09T16:19:34.1794199Z y3 = F.linear(x, w3[index]) 2025-09-09T16:19:34.1794361Z 2025-09-09T16:19:34.1794504Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1795278Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1795979Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1796881Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 103, in forward 2025-09-09T16:19:34.1797525Z cur_out = F.linear(y1 * y3, y2) 2025-09-09T16:19:34.1797707Z 2025-09-09T16:19:34.1797851Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1798539Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1799239Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1799930Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 99, in forward 2025-09-09T16:19:34.1800579Z y1 = F.silu(F.linear(x, w1[index])) 2025-09-09T16:19:34.1800766Z 2025-09-09T16:19:34.1800910Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1801645Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1802346Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1803040Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 100, in forward 2025-09-09T16:19:34.1803684Z y3 = F.linear(x, w3[index]) 2025-09-09T16:19:34.1803845Z 2025-09-09T16:19:34.1803989Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1804675Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1805378Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1806078Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 103, in forward 2025-09-09T16:19:34.1806722Z cur_out = F.linear(y1 * y3, y2) 2025-09-09T16:19:34.1806901Z 2025-09-09T16:19:34.1807048Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1807743Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1808442Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1809142Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 95, in forward 2025-09-09T16:19:34.1809788Z w2 = self.w2[expert_indices] 2025-09-09T16:19:34.1809957Z 2025-09-09T16:19:34.1810101Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1810792Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1811482Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1812194Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 103, in forward 2025-09-09T16:19:34.1812834Z cur_out = F.linear(y1 * y3, y2) 2025-09-09T16:19:34.1813015Z 2025-09-09T16:19:34.1813123Z cudagraph partition due to non gpu ops 2025-09-09T16:19:34.1813463Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1814141Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1814840Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1815539Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 103, in forward 2025-09-09T16:19:34.1816170Z cur_out = F.linear(y1 * y3, y2) 2025-09-09T16:19:34.1816400Z 2025-09-09T16:19:34.1816514Z cudagraph partition due to non gpu ops 2025-09-09T16:19:34.1816847Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1817583Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1818320Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1819016Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 108, in forward 2025-09-09T16:19:34.1819718Z (torch.cat(outs, dim=0) * expert_weights.view(-1, 1)) 2025-09-09T16:19:34.1819955Z 2025-09-09T16:19:34.1820100Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1820784Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 40, in forward 2025-09-09T16:19:34.1821476Z out = self.experts(x, expert_indices, scores, self.top_k) 2025-09-09T16:19:34.1822183Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 108, in forward 2025-09-09T16:19:34.1822921Z (torch.cat(outs, dim=0) * expert_weights.view(-1, 1)) 2025-09-09T16:19:34.1823156Z 2025-09-09T16:19:34.1823466Z PASSED 2025-09-09T16:19:34.1824092Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_base_cpu_1_multiple_tokens cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1825128Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 33, in forward 2025-09-09T16:19:34.1825771Z scores = self.router(x) # [T, E] 2025-09-09T16:19:34.1825956Z 2025-09-09T16:19:34.1826066Z cudagraph partition due to non gpu ops 2025-09-09T16:19:34.1826406Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1827091Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 35, in forward 2025-09-09T16:19:34.1827739Z scores, expert_indices = torch.topk( 2025-09-09T16:19:34.1827945Z 2025-09-09T16:19:34.1828091Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1828778Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 35, in forward 2025-09-09T16:19:34.1829423Z scores, expert_indices = torch.topk( 2025-09-09T16:19:34.1829618Z 2025-09-09T16:19:34.1829770Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1830449Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 38, in forward 2025-09-09T16:19:34.1831172Z scores /= scores.sum(dim=-1, keepdim=True).to(x.dtype) # [T, A] 2025-09-09T16:19:34.1831443Z 2025-09-09T16:19:34.1831559Z cudagraph partition due to non gpu ops 2025-09-09T16:19:34.1831893Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1832596Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 125, in forward 2025-09-09T16:19:34.1833314Z expert_indices.view(-1) + 1, minlength=self.num_experts + 1 2025-09-09T16:19:34.1833585Z 2025-09-09T16:19:34.1833737Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1834421Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 117, in forward 2025-09-09T16:19:34.1835142Z ordered_token_activations = expert_indices.view(-1).argsort( 2025-09-09T16:19:34.1835411Z 2025-09-09T16:19:34.1835564Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:34.1836345Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 124, in forward 2025-09-09T16:19:38.2381327Z num_tokens_per_expert = torch.bincount( 2025-09-09T16:19:38.2381831Z 2025-09-09T16:19:38.2381997Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2382796Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 124, in forward 2025-09-09T16:19:38.2383481Z num_tokens_per_expert = torch.bincount( 2025-09-09T16:19:38.2383757Z 2025-09-09T16:19:38.2383912Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2384603Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 134, in forward 2025-09-09T16:19:38.2385322Z cum_tokens_per_expert = num_tokens_per_expert.cumsum(0).to( 2025-09-09T16:19:38.2385586Z 2025-09-09T16:19:38.2385696Z cudagraph partition due to non gpu ops 2025-09-09T16:19:38.2386005Z cudagraph partition due to non gpu ops 2025-09-09T16:19:38.2386343Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2387037Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 134, in forward 2025-09-09T16:19:38.2387763Z cum_tokens_per_expert = num_tokens_per_expert.cumsum(0).to( 2025-09-09T16:19:38.2388104Z 2025-09-09T16:19:38.2388250Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2388941Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 121, in forward 2025-09-09T16:19:38.2389676Z ordered_token_activations.div(top_k).floor().to(torch.int64) 2025-09-09T16:19:38.2389948Z 2025-09-09T16:19:38.2390094Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2390890Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2391639Z tokens_grouped_by_expert = [ 2025-09-09T16:19:38.2392293Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:38.2393007Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:38.2393242Z 2025-09-09T16:19:38.2393389Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2394201Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2394963Z final_out = final_out.scatter_add( 2025-09-09T16:19:38.2395165Z 2025-09-09T16:19:38.2395310Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2396113Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2396949Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:38.2397129Z 2025-09-09T16:19:38.2397277Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2398071Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2398828Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:38.2398990Z 2025-09-09T16:19:38.2399144Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2399938Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2400704Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2400933Z 2025-09-09T16:19:38.2401076Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2401878Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2402638Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2402908Z 2025-09-09T16:19:38.2403055Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2403900Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2404683Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:38.2404872Z 2025-09-09T16:19:38.2405015Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2405813Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2406567Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:38.2406727Z 2025-09-09T16:19:38.2406877Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2407672Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2408444Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2408650Z 2025-09-09T16:19:38.2408839Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2409639Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2410407Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2410641Z 2025-09-09T16:19:38.2410808Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2411612Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2412360Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:38.2412547Z 2025-09-09T16:19:38.2412692Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2413500Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2414242Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:38.2414409Z 2025-09-09T16:19:38.2414556Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2415349Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2416114Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2416326Z 2025-09-09T16:19:38.2416474Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2417269Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2418038Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2418244Z 2025-09-09T16:19:38.2418388Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2419194Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2419947Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:38.2420131Z 2025-09-09T16:19:38.2420278Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2421075Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2421812Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:38.2421978Z 2025-09-09T16:19:38.2422124Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2422920Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2423732Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2423943Z 2025-09-09T16:19:38.2424088Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2424959Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2425726Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2425929Z 2025-09-09T16:19:38.2426078Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2426864Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2427611Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:38.2427796Z 2025-09-09T16:19:38.2427937Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2428731Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2429481Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:38.2429683Z 2025-09-09T16:19:38.2429826Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2430628Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2431439Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2431648Z 2025-09-09T16:19:38.2431790Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2432590Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2433350Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2433560Z 2025-09-09T16:19:38.2433701Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2434495Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2435253Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:38.2435433Z 2025-09-09T16:19:38.2435585Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2436468Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2437210Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:38.2437367Z 2025-09-09T16:19:38.2437508Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2438312Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2439080Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2439287Z 2025-09-09T16:19:38.2439433Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:38.2440232Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:38.2441046Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:38.2441256Z 2025-09-09T16:19:38.2441399Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2969344Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2970276Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:41.2970534Z 2025-09-09T16:19:41.2970696Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2971675Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2972710Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:41.2972950Z 2025-09-09T16:19:41.2973228Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2974428Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2975376Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.2975658Z 2025-09-09T16:19:41.2975822Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2976808Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2977743Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.2977982Z 2025-09-09T16:19:41.2978145Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2979095Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2980034Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:41.2980221Z 2025-09-09T16:19:41.2980370Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2981171Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2981919Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:41.2982080Z 2025-09-09T16:19:41.2982225Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2983077Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2983833Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.2984050Z 2025-09-09T16:19:41.2984193Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2984994Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2985759Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.2985970Z 2025-09-09T16:19:41.2986080Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2986382Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2986682Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2986974Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2987276Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2987581Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2987874Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2988177Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2988506Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2989312Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2990094Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:41.2990332Z 2025-09-09T16:19:41.2990482Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2991282Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2992057Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:41.2992287Z 2025-09-09T16:19:41.2992393Z cudagraph partition due to non gpu ops 2025-09-09T16:19:41.2992771Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2993572Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2994398Z tokens_grouped_by_expert = [ 2025-09-09T16:19:41.2995082Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:41.2995793Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:41.2996816Z 2025-09-09T16:19:41.2996982Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.2997786Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.2998541Z tokens_grouped_by_expert = [ 2025-09-09T16:19:41.2999184Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:41.2999894Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:41.3000127Z 2025-09-09T16:19:41.3000280Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3001079Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3001886Z final_out = final_out.scatter_add( 2025-09-09T16:19:41.3002081Z 2025-09-09T16:19:41.3002231Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3003033Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3003781Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:41.3003969Z 2025-09-09T16:19:41.3004119Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3004921Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3005668Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:41.3005827Z 2025-09-09T16:19:41.3005978Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3006778Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3007550Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3007759Z 2025-09-09T16:19:41.3007911Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3008711Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3009486Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3009696Z 2025-09-09T16:19:41.3009840Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3010650Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3011408Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:41.3011598Z 2025-09-09T16:19:41.3011745Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3012556Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3013300Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:41.3013467Z 2025-09-09T16:19:41.3013614Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3014421Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3015182Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3015399Z 2025-09-09T16:19:41.3015555Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3016405Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3017228Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3017436Z 2025-09-09T16:19:41.3017630Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3018427Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3019187Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:41.3019368Z 2025-09-09T16:19:41.3019523Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3020331Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3021080Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:41.3021242Z 2025-09-09T16:19:41.3021389Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3022197Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3023066Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3023360Z 2025-09-09T16:19:41.3023552Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3024341Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3025095Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3025304Z 2025-09-09T16:19:41.3025446Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3026235Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3026988Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:41.3027168Z 2025-09-09T16:19:41.3027322Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3028112Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3028847Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:41.3029003Z 2025-09-09T16:19:41.3029145Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:41.3029945Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:41.3030706Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:41.3030909Z 2025-09-09T16:19:41.3031051Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1965484Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1966433Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1966669Z 2025-09-09T16:19:44.1966821Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1967643Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1968399Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.1968587Z 2025-09-09T16:19:44.1968740Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1969538Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1970285Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.1970446Z 2025-09-09T16:19:44.1970799Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1971673Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1972445Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1972655Z 2025-09-09T16:19:44.1972861Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1973661Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1974438Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1974646Z 2025-09-09T16:19:44.1974788Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1975582Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1976330Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.1976521Z 2025-09-09T16:19:44.1976662Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1977536Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1978268Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.1978425Z 2025-09-09T16:19:44.1978573Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1979363Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1980139Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1980347Z 2025-09-09T16:19:44.1987999Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1988867Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1989643Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1989854Z 2025-09-09T16:19:44.1990006Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1990808Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1991551Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.1991743Z 2025-09-09T16:19:44.1991890Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1992690Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1993425Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.1993585Z 2025-09-09T16:19:44.1993739Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1994531Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1995308Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1995517Z 2025-09-09T16:19:44.1995673Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1996518Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1997282Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.1997491Z 2025-09-09T16:19:44.1997634Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.1998425Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.1999176Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.1999434Z 2025-09-09T16:19:44.1999582Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2000580Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2001509Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.2001684Z 2025-09-09T16:19:44.2001846Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2002790Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2003704Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.2003940Z 2025-09-09T16:19:44.2004103Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2005052Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2005979Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.2006258Z 2025-09-09T16:19:44.2006380Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2006720Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2007055Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2007386Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2007717Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2008042Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2008372Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2008699Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2009078Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2010032Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2010975Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:44.2011235Z 2025-09-09T16:19:44.2011404Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2012357Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2013310Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:44.2013572Z 2025-09-09T16:19:44.2013684Z cudagraph partition due to non gpu ops 2025-09-09T16:19:44.2014064Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2015024Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2015920Z final_out = final_out.scatter_add( 2025-09-09T16:19:44.2016136Z 2025-09-09T16:19:44.2016301Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2017246Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2018142Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.2018344Z 2025-09-09T16:19:44.2018509Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2019456Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2020344Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.2020518Z 2025-09-09T16:19:44.2020673Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2021623Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2022532Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.2022765Z 2025-09-09T16:19:44.2022978Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2023971Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2024878Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.2025163Z 2025-09-09T16:19:44.2025320Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2026270Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2027158Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.2027358Z 2025-09-09T16:19:44.2027520Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2028465Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2029354Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.2029530Z 2025-09-09T16:19:44.2029700Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2030712Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2031627Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.2031861Z 2025-09-09T16:19:44.2032018Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2032976Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2033899Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:44.2034133Z 2025-09-09T16:19:44.2034288Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2035239Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2036232Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:44.2036425Z 2025-09-09T16:19:44.2036571Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:44.2037363Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:44.2038091Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:44.2038258Z 2025-09-09T16:19:46.7672238Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7673351Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7674135Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7674359Z 2025-09-09T16:19:46.7674527Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7675653Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7676507Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7676715Z 2025-09-09T16:19:46.7676872Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7677790Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7678821Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:46.7679059Z 2025-09-09T16:19:46.7679203Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7679993Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7680734Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:46.7681042Z 2025-09-09T16:19:46.7681185Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7682049Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7682861Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7683077Z 2025-09-09T16:19:46.7683222Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7684008Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7684760Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7684971Z 2025-09-09T16:19:46.7685112Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7685897Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7686703Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:46.7686953Z 2025-09-09T16:19:46.7687105Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7687891Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7688633Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:46.7688788Z 2025-09-09T16:19:46.7688930Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7689716Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7690473Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7690676Z 2025-09-09T16:19:46.7690822Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7691617Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7692372Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7692584Z 2025-09-09T16:19:46.7692729Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7693524Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7694266Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:46.7694455Z 2025-09-09T16:19:46.7694603Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7695391Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7696127Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:46.7696287Z 2025-09-09T16:19:46.7696437Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7697222Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7697985Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7698190Z 2025-09-09T16:19:46.7698332Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7699122Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7699883Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7700085Z 2025-09-09T16:19:46.7700231Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7701021Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7701812Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:46.7702006Z 2025-09-09T16:19:46.7702148Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7703019Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7703755Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:46.7703920Z 2025-09-09T16:19:46.7704063Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7704882Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7705638Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7705847Z 2025-09-09T16:19:46.7705994Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7706778Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7707583Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7707793Z 2025-09-09T16:19:46.7707947Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7708741Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7709491Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:46.7709672Z 2025-09-09T16:19:46.7709816Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7710616Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7711355Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:46.7711511Z 2025-09-09T16:19:46.7711656Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7712463Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7713220Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7713430Z 2025-09-09T16:19:46.7713575Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7714370Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7715133Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7715346Z 2025-09-09T16:19:46.7715455Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7715757Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7716063Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7716446Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7716750Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7717088Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7717381Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7717677Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7718006Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7718801Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7719575Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:46.7719809Z 2025-09-09T16:19:46.7719953Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7720749Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7721527Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:46.7721817Z 2025-09-09T16:19:46.7721926Z cudagraph partition due to non gpu ops 2025-09-09T16:19:46.7722260Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7723110Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7723913Z tokens_grouped_by_expert = [ 2025-09-09T16:19:46.7724557Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:46.7725275Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:46.7725503Z 2025-09-09T16:19:46.7725648Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7726454Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7727216Z final_out = final_out.scatter_add( 2025-09-09T16:19:46.7727414Z 2025-09-09T16:19:46.7727559Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7728413Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7729159Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:46.7729352Z 2025-09-09T16:19:46.7729496Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7730314Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7731051Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:46.7731224Z 2025-09-09T16:19:46.7731367Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:46.7732170Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:46.7732941Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:46.7733152Z 2025-09-09T16:19:49.3490941Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3491881Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3492992Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3493214Z 2025-09-09T16:19:49.3493359Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3494163Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3494902Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3495087Z 2025-09-09T16:19:49.3495228Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3496022Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3496767Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3496949Z 2025-09-09T16:19:49.3497092Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3497890Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3498744Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3498954Z 2025-09-09T16:19:49.3499122Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3500139Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3500928Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3501367Z 2025-09-09T16:19:49.3501516Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3502418Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3503174Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3503434Z 2025-09-09T16:19:49.3503578Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3504383Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3505121Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3505277Z 2025-09-09T16:19:49.3505429Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3506225Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3506999Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3507209Z 2025-09-09T16:19:49.3507359Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3508233Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3508995Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3509199Z 2025-09-09T16:19:49.3509342Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3510137Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3510883Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3511063Z 2025-09-09T16:19:49.3511206Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3512010Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3512745Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3512908Z 2025-09-09T16:19:49.3513052Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3513860Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3514619Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3514824Z 2025-09-09T16:19:49.3514971Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3515760Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3516605Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3516809Z 2025-09-09T16:19:49.3516963Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3517754Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3518502Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3518685Z 2025-09-09T16:19:49.3518829Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3519628Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3520365Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3520520Z 2025-09-09T16:19:49.3520662Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3521458Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3522215Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3522486Z 2025-09-09T16:19:49.3522630Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3523469Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3524265Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3524470Z 2025-09-09T16:19:49.3524622Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3525411Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3526171Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3526353Z 2025-09-09T16:19:49.3526505Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3527294Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3528097Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3528301Z 2025-09-09T16:19:49.3528447Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3529250Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3530016Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3530220Z 2025-09-09T16:19:49.3530366Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3531167Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3531928Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3532141Z 2025-09-09T16:19:49.3532285Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3533088Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3533838Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3534017Z 2025-09-09T16:19:49.3534176Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3534977Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3535725Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3535881Z 2025-09-09T16:19:49.3536037Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3536828Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3537589Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3537795Z 2025-09-09T16:19:49.3537940Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3538788Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3539564Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3539772Z 2025-09-09T16:19:49.3539884Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3540195Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3540489Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3540788Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3541075Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3541369Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3541656Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3541995Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3542793Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3543629Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:49.3543899Z 2025-09-09T16:19:49.3544042Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3544873Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3545659Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:49.3545888Z 2025-09-09T16:19:49.3546009Z cudagraph partition due to non gpu ops 2025-09-09T16:19:49.3546338Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3547132Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3547884Z tokens_grouped_by_expert = [ 2025-09-09T16:19:49.3548528Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:49.3549281Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:49.3549510Z 2025-09-09T16:19:49.3549655Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3550458Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3551209Z final_out = final_out.scatter_add( 2025-09-09T16:19:49.3551406Z 2025-09-09T16:19:49.3551547Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3552343Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3553088Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3553273Z 2025-09-09T16:19:49.3553421Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3554219Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3554965Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3555123Z 2025-09-09T16:19:49.3555276Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3556062Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3556914Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3557122Z 2025-09-09T16:19:49.3557264Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3558115Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3558891Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3559099Z 2025-09-09T16:19:49.3559247Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3560048Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3560795Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3560989Z 2025-09-09T16:19:49.3561133Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3561935Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3562677Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3562835Z 2025-09-09T16:19:49.3562987Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3564100Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3565434Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3565746Z 2025-09-09T16:19:49.3565899Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3566753Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3567518Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3567721Z 2025-09-09T16:19:49.3567861Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3568715Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3569466Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3569644Z 2025-09-09T16:19:49.3569792Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3570595Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3571395Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3571558Z 2025-09-09T16:19:49.3571704Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3572503Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3573269Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3573480Z 2025-09-09T16:19:49.3573626Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3574417Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3575182Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3575385Z 2025-09-09T16:19:49.3575535Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3576332Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3577081Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3577260Z 2025-09-09T16:19:49.3577400Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3578202Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3578944Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3579106Z 2025-09-09T16:19:49.3579248Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3580048Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3580807Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3581028Z 2025-09-09T16:19:49.3581173Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3581974Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3582736Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3582943Z 2025-09-09T16:19:49.3583100Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3583891Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3584645Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3584829Z 2025-09-09T16:19:49.3584981Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3585835Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3586625Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3586789Z 2025-09-09T16:19:49.3586971Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3587776Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3588596Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3588801Z 2025-09-09T16:19:49.3588942Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3589745Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3590510Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3590731Z 2025-09-09T16:19:49.3590873Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3591731Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3592474Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3592652Z 2025-09-09T16:19:49.3592802Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3593592Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3594340Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3594497Z 2025-09-09T16:19:49.3594649Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3595438Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3596249Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3596458Z 2025-09-09T16:19:49.3596604Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3597408Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3598183Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3598386Z 2025-09-09T16:19:49.3598543Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3599383Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3600125Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3600317Z 2025-09-09T16:19:49.3600460Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3601261Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3602001Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3602156Z 2025-09-09T16:19:49.3602308Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3603098Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3603860Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3604064Z 2025-09-09T16:19:49.3604210Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3605004Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3605766Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3606026Z 2025-09-09T16:19:49.3606171Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3607015Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3607770Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:49.3607985Z 2025-09-09T16:19:49.3608128Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3608927Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3609663Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:49.3609822Z 2025-09-09T16:19:49.3609964Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:49.3610762Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:49.3611521Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:49.3611725Z 2025-09-09T16:19:49.3611876Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0716831Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0717898Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0718196Z 2025-09-09T16:19:52.0718328Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0718743Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0726759Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0727076Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0727378Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0727679Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0727976Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0728286Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0728645Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0729490Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0730304Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:52.0730541Z 2025-09-09T16:19:52.0730689Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0731504Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0732297Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:52.0732537Z 2025-09-09T16:19:52.0732650Z cudagraph partition due to non gpu ops 2025-09-09T16:19:52.0733012Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0733820Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0734595Z tokens_grouped_by_expert = [ 2025-09-09T16:19:52.0735263Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:52.0735990Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:52.0736221Z 2025-09-09T16:19:52.0736380Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0737184Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0737938Z tokens_grouped_by_expert = [ 2025-09-09T16:19:52.0738589Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:52.0739291Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:52.0739743Z 2025-09-09T16:19:52.0739902Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0740783Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0741618Z final_out = final_out.scatter_add( 2025-09-09T16:19:52.0741813Z 2025-09-09T16:19:52.0741965Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0742774Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0743536Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:52.0743719Z 2025-09-09T16:19:52.0743867Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0744678Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0745435Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:52.0745685Z 2025-09-09T16:19:52.0745848Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0746660Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0747432Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0747655Z 2025-09-09T16:19:52.0747807Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0748613Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0749390Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0749600Z 2025-09-09T16:19:52.0749781Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0750610Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0751379Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:52.0751563Z 2025-09-09T16:19:52.0751731Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0752532Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0753279Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:52.0753443Z 2025-09-09T16:19:52.0753587Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0754390Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0755156Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0755368Z 2025-09-09T16:19:52.0755515Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0756401Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0757170Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0757383Z 2025-09-09T16:19:52.0757528Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0758324Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0759070Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:52.0759254Z 2025-09-09T16:19:52.0759409Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0760224Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0761063Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:52.0761226Z 2025-09-09T16:19:52.0761379Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0762263Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0763034Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0763241Z 2025-09-09T16:19:52.0763386Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0764860Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0765683Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0765892Z 2025-09-09T16:19:52.0766037Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0766841Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0767715Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:52.0767909Z 2025-09-09T16:19:52.0768055Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0768859Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0769599Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:52.0769762Z 2025-09-09T16:19:52.0769912Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0770709Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0771475Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0771683Z 2025-09-09T16:19:52.0771842Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0772642Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0773411Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0773620Z 2025-09-09T16:19:52.0773768Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0774573Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0775327Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:52.0775509Z 2025-09-09T16:19:52.0775658Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0776467Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0777213Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:52.0777382Z 2025-09-09T16:19:52.0777530Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0778341Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0779106Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0779317Z 2025-09-09T16:19:52.0779469Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0780312Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0781077Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:52.0781284Z 2025-09-09T16:19:52.0781439Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0782236Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0783069Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:52.0783254Z 2025-09-09T16:19:52.0783465Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:52.0784323Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:52.0785077Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:52.0785235Z 2025-09-09T16:19:52.0785381Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6040222Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6041344Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6041561Z 2025-09-09T16:19:54.6041723Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6042526Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6043457Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6043666Z 2025-09-09T16:19:54.6043821Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6044610Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6045353Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:54.6045533Z 2025-09-09T16:19:54.6045675Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6046467Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6047203Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:54.6047361Z 2025-09-09T16:19:54.6047502Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6048294Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6049056Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6049266Z 2025-09-09T16:19:54.6049406Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6050185Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6050947Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6051152Z 2025-09-09T16:19:54.6051278Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6051614Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6051909Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6052198Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6052486Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6052772Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6053070Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6053404Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6054194Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6054980Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:54.6055201Z 2025-09-09T16:19:54.6055341Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6056135Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6056912Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:54.6057221Z 2025-09-09T16:19:54.6057328Z cudagraph partition due to non gpu ops 2025-09-09T16:19:54.6057667Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6058523Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6059326Z tokens_grouped_by_expert = [ 2025-09-09T16:19:54.6059978Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:54.6060670Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:54.6060898Z 2025-09-09T16:19:54.6061049Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6061840Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6062599Z final_out = final_out.scatter_add( 2025-09-09T16:19:54.6062793Z 2025-09-09T16:19:54.6062950Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6063961Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6064717Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:54.6064896Z 2025-09-09T16:19:54.6065043Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6065835Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6066570Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:54.6066731Z 2025-09-09T16:19:54.6066872Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6067666Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6068421Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6068631Z 2025-09-09T16:19:54.6068780Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6069575Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6070327Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6070533Z 2025-09-09T16:19:54.6070680Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6071458Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6072200Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:54.6072377Z 2025-09-09T16:19:54.6072524Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6073311Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6074055Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:54.6074215Z 2025-09-09T16:19:54.6074356Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6075155Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6075914Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6076118Z 2025-09-09T16:19:54.6076331Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6077121Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6077872Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6078208Z 2025-09-09T16:19:54.6078348Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6079192Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6079934Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:54.6080168Z 2025-09-09T16:19:54.6080317Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6081103Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6081889Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:54.6082044Z 2025-09-09T16:19:54.6082189Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6082974Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6083739Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6083941Z 2025-09-09T16:19:54.6084081Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6084940Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6085699Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6085901Z 2025-09-09T16:19:54.6086045Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6086837Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6087582Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:54.6087772Z 2025-09-09T16:19:54.6087915Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6088709Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6089447Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:54.6089608Z 2025-09-09T16:19:54.6089758Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6090548Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6091312Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6091520Z 2025-09-09T16:19:54.6091673Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6092464Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6093226Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6093431Z 2025-09-09T16:19:54.6093581Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6094372Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6095119Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:54.6095301Z 2025-09-09T16:19:54.6095447Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6096238Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6096993Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:54.6097149Z 2025-09-09T16:19:54.6097296Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:54.6098087Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:54.6098845Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:54.6099101Z 2025-09-09T16:19:54.6099242Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1056797Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1057894Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1058115Z 2025-09-09T16:19:57.1058281Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1059244Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1060148Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:57.1060356Z 2025-09-09T16:19:57.1060519Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1061481Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1062382Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:57.1062653Z 2025-09-09T16:19:57.1062817Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1063946Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1064706Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1064921Z 2025-09-09T16:19:57.1065066Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1065851Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1066608Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1066815Z 2025-09-09T16:19:57.1066967Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1067753Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1068500Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:57.1068681Z 2025-09-09T16:19:57.1068826Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1069615Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1070353Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:57.1070509Z 2025-09-09T16:19:57.1070650Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1071444Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1072197Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1072414Z 2025-09-09T16:19:57.1072555Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1073370Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1074150Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1074361Z 2025-09-09T16:19:57.1074471Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1074772Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1075068Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1075362Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1075654Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1075950Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1076338Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1076673Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1077456Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1078342Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:57.1078625Z 2025-09-09T16:19:57.1078776Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1079618Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1080403Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:57.1080635Z 2025-09-09T16:19:57.1080745Z cudagraph partition due to non gpu ops 2025-09-09T16:19:57.1081084Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1081875Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1082634Z tokens_grouped_by_expert = [ 2025-09-09T16:19:57.1083341Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:57.1084104Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:57.1084338Z 2025-09-09T16:19:57.1084488Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1085276Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1086031Z tokens_grouped_by_expert = [ 2025-09-09T16:19:57.1086678Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:57.1087377Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:57.1087616Z 2025-09-09T16:19:57.1087761Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1088553Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1089320Z final_out = final_out.scatter_add( 2025-09-09T16:19:57.1089512Z 2025-09-09T16:19:57.1089663Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1090463Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1091214Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:57.1091393Z 2025-09-09T16:19:57.1091539Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1092339Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1093086Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:57.1093252Z 2025-09-09T16:19:57.1093395Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1094197Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1094960Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1095184Z 2025-09-09T16:19:57.1095327Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1096126Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1096882Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1097094Z 2025-09-09T16:19:57.1097241Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1098030Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1099381Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:57.1099559Z 2025-09-09T16:19:57.1099710Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1100592Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1101335Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:57.1101494Z 2025-09-09T16:19:57.1101638Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1102425Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1103185Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1103388Z 2025-09-09T16:19:57.1103528Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1104315Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1105074Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1105338Z 2025-09-09T16:19:57.1105481Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1106277Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1107011Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:57.1107203Z 2025-09-09T16:19:57.1107347Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1108130Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1108874Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:57.1109033Z 2025-09-09T16:19:57.1109180Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1109972Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1110743Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1110946Z 2025-09-09T16:19:57.1111089Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1111883Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1112640Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:57.1112847Z 2025-09-09T16:19:57.1112988Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1113777Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1114522Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:57.1114710Z 2025-09-09T16:19:57.1114856Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:57.1115656Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:57.1116441Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:57.1116605Z 2025-09-09T16:19:57.1116752Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6089196Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6090220Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6090436Z 2025-09-09T16:19:59.6090599Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6091397Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6092312Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6092521Z 2025-09-09T16:19:59.6092743Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6093601Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6094352Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:59.6094534Z 2025-09-09T16:19:59.6094678Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6095474Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6096211Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:59.6096379Z 2025-09-09T16:19:59.6096521Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6097327Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6098203Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6098406Z 2025-09-09T16:19:59.6098562Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6099347Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6100104Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6100311Z 2025-09-09T16:19:59.6100466Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6101261Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6102007Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:59.6102196Z 2025-09-09T16:19:59.6102339Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6103134Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6103880Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:59.6104039Z 2025-09-09T16:19:59.6104183Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6104981Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6105786Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6106001Z 2025-09-09T16:19:59.6106143Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6106939Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6107698Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6107903Z 2025-09-09T16:19:59.6108023Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6108322Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6108629Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6108919Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6109211Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6109499Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6109831Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6110628Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6111399Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:19:59.6111629Z 2025-09-09T16:19:59.6111770Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6112613Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6113433Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:19:59.6113664Z 2025-09-09T16:19:59.6113817Z cudagraph partition due to non gpu ops 2025-09-09T16:19:59.6114146Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6114987Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6115726Z tokens_grouped_by_expert = [ 2025-09-09T16:19:59.6116461Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:19:59.6117158Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:19:59.6117383Z 2025-09-09T16:19:59.6117529Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6118321Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6119111Z final_out = final_out.scatter_add( 2025-09-09T16:19:59.6119307Z 2025-09-09T16:19:59.6119458Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6120247Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6120984Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:59.6121171Z 2025-09-09T16:19:59.6121317Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6122106Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6122846Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:59.6123005Z 2025-09-09T16:19:59.6123153Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6123944Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6124714Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6124924Z 2025-09-09T16:19:59.6125065Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6125864Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6126624Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6126830Z 2025-09-09T16:19:59.6126981Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6127774Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6128509Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:59.6128703Z 2025-09-09T16:19:59.6128848Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6129645Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6130378Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:59.6130540Z 2025-09-09T16:19:59.6130697Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6131485Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6132248Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6132455Z 2025-09-09T16:19:59.6132605Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6133445Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6134253Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6134463Z 2025-09-09T16:19:59.6134645Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6135493Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6136243Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:59.6136424Z 2025-09-09T16:19:59.6136568Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6137369Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6138099Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:59.6138270Z 2025-09-09T16:19:59.6138412Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6139207Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6140008Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6140213Z 2025-09-09T16:19:59.6140363Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6141170Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6141932Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6150003Z 2025-09-09T16:19:59.6150200Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6151014Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6151773Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:19:59.6151954Z 2025-09-09T16:19:59.6152112Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6152911Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6153648Z y3 = F.linear(cur_x, w3) 2025-09-09T16:19:59.6153810Z 2025-09-09T16:19:59.6153953Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:19:59.6154748Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:19:59.6155512Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:19:59.6155716Z 2025-09-09T16:19:59.6155859Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1262072Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1263150Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1263391Z 2025-09-09T16:20:02.1263577Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1264779Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1265697Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:02.1265939Z 2025-09-09T16:20:02.1266091Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1267071Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1267813Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:02.1267980Z 2025-09-09T16:20:02.1268436Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1269322Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1270089Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1270389Z 2025-09-09T16:20:02.1270534Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1271326Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1272090Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1272300Z 2025-09-09T16:20:02.1272455Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1273245Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1274003Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:02.1274185Z 2025-09-09T16:20:02.1274330Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1275220Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1275962Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:02.1276118Z 2025-09-09T16:20:02.1276350Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1277197Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1277951Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1278161Z 2025-09-09T16:20:02.1278303Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1279094Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1279848Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1280058Z 2025-09-09T16:20:02.1280205Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1280992Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1281742Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:02.1281921Z 2025-09-09T16:20:02.1282075Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1282860Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1283599Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:02.1283761Z 2025-09-09T16:20:02.1283902Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1284696Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1285458Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1285663Z 2025-09-09T16:20:02.1285807Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1286596Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1287354Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1287564Z 2025-09-09T16:20:02.1287672Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1287975Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1288272Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1288568Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1288864Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1289213Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1289503Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1289888Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1290716Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1291503Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:20:02.1291727Z 2025-09-09T16:20:02.1291878Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1292664Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1293445Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:20:02.1293668Z 2025-09-09T16:20:02.1293774Z cudagraph partition due to non gpu ops 2025-09-09T16:20:02.1294118Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1294918Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1295706Z tokens_grouped_by_expert = [ 2025-09-09T16:20:02.1296383Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:20:02.1297105Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:20:02.1297338Z 2025-09-09T16:20:02.1297486Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1298279Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1299021Z tokens_grouped_by_expert = [ 2025-09-09T16:20:02.1299666Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:20:02.1300357Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:20:02.1300596Z 2025-09-09T16:20:02.1300747Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1301552Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1302297Z final_out = final_out.scatter_add( 2025-09-09T16:20:02.1302497Z 2025-09-09T16:20:02.1302641Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1303427Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1304183Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:02.1304365Z 2025-09-09T16:20:02.1304518Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1305306Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1306051Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:02.1306208Z 2025-09-09T16:20:02.1306356Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1307154Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1307912Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1308115Z 2025-09-09T16:20:02.1308256Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1309044Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1309797Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1310058Z 2025-09-09T16:20:02.1310202Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1311044Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1311817Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:02.1312003Z 2025-09-09T16:20:02.1312147Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1312927Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1313661Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:02.1313819Z 2025-09-09T16:20:02.1313967Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1314749Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1315508Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1315713Z 2025-09-09T16:20:02.1315902Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1316784Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1317536Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:02.1317742Z 2025-09-09T16:20:02.1317884Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1318674Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1319413Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:02.1319597Z 2025-09-09T16:20:02.1319740Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:02.1320537Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:02.1321272Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:02.1321431Z 2025-09-09T16:20:02.1321580Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6405772Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6407278Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6407662Z 2025-09-09T16:20:04.6407927Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6409193Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6409954Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6410167Z 2025-09-09T16:20:04.6410310Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6411108Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6411852Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:04.6412042Z 2025-09-09T16:20:04.6412187Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6412985Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6413723Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:04.6413891Z 2025-09-09T16:20:04.6414031Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6414815Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6415789Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6415995Z 2025-09-09T16:20:04.6416146Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6417097Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6417860Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6418066Z 2025-09-09T16:20:04.6418208Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6419000Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6419749Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:04.6419928Z 2025-09-09T16:20:04.6420073Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6420868Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6421606Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:04.6421847Z 2025-09-09T16:20:04.6421996Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6422795Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6423548Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6423757Z 2025-09-09T16:20:04.6423905Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6424687Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6425449Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6425655Z 2025-09-09T16:20:04.6425809Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6426597Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6427344Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:04.6427526Z 2025-09-09T16:20:04.6427671Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6428465Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6429207Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:04.6429367Z 2025-09-09T16:20:04.6429512Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6430306Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6431055Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6431269Z 2025-09-09T16:20:04.6431411Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6432212Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6432971Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6433183Z 2025-09-09T16:20:04.6433326Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6434106Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6434854Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:04.6435032Z 2025-09-09T16:20:04.6435186Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6435966Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6436841Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:04.6437001Z 2025-09-09T16:20:04.6437231Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6438067Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6438826Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6439029Z 2025-09-09T16:20:04.6439171Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6439965Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6440713Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6440927Z 2025-09-09T16:20:04.6441035Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6441341Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6441632Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6441927Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6442261Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6442553Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6442843Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6443174Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6443970Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6444743Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:20:04.6444964Z 2025-09-09T16:20:04.6445110Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6445896Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6446676Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:20:04.6446898Z 2025-09-09T16:20:04.6447012Z cudagraph partition due to non gpu ops 2025-09-09T16:20:04.6447341Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6448138Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 155, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6448877Z tokens_grouped_by_expert = [ 2025-09-09T16:20:04.6449525Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 156, in 2025-09-09T16:20:04.6450221Z x[indices] for indices in token_indices_per_expert 2025-09-09T16:20:04.6450457Z 2025-09-09T16:20:04.6450600Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6451398Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 184, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6452158Z final_out = final_out.scatter_add( 2025-09-09T16:20:04.6452357Z 2025-09-09T16:20:04.6452509Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6453300Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6454039Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:04.6454224Z 2025-09-09T16:20:04.6454372Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6455158Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6455905Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:04.6456062Z 2025-09-09T16:20:04.6456217Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6457058Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6457877Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6458108Z 2025-09-09T16:20:04.6458313Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6459112Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6459870Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6460075Z 2025-09-09T16:20:04.6460216Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6461014Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6461757Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:20:04.6461951Z 2025-09-09T16:20:04.6462091Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6462890Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6463668Z y3 = F.linear(cur_x, w3) 2025-09-09T16:20:04.6464018Z 2025-09-09T16:20:04.6464173Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:20:04.6464954Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:20:04.6465712Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:20:04.6465917Z 2025-09-09T16:20:04.6466070Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1969662Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1972520Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.1972826Z 2025-09-09T16:31:41.1973024Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1974037Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1974967Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:31:41.1975191Z 2025-09-09T16:31:41.1975364Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1976341Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1977262Z y3 = F.linear(cur_x, w3) 2025-09-09T16:31:41.1977461Z 2025-09-09T16:31:41.1977633Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1978621Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1979600Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.1979857Z 2025-09-09T16:31:41.1980036Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1981025Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1981983Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.1982237Z 2025-09-09T16:31:41.1982408Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1983395Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1984356Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:31:41.1984576Z 2025-09-09T16:31:41.1985130Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1986230Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1987148Z y3 = F.linear(cur_x, w3) 2025-09-09T16:31:41.1987348Z 2025-09-09T16:31:41.1987624Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1988604Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1989536Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.1989791Z 2025-09-09T16:31:41.1989962Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1990930Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1991873Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.1992124Z 2025-09-09T16:31:41.1992304Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1993393Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1994329Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:31:41.1994551Z 2025-09-09T16:31:41.1994725Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1995715Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1996738Z y3 = F.linear(cur_x, w3) 2025-09-09T16:31:41.1996928Z 2025-09-09T16:31:41.1997099Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.1998093Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.1999046Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.1999308Z 2025-09-09T16:31:41.1999483Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2000519Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2001480Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.2001739Z 2025-09-09T16:31:41.2001911Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2002898Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2003842Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:31:41.2004060Z 2025-09-09T16:31:41.2004242Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2005228Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2006170Z y3 = F.linear(cur_x, w3) 2025-09-09T16:31:41.2006360Z 2025-09-09T16:31:41.2006534Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2007534Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2008508Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.2008757Z 2025-09-09T16:31:41.2008928Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2009904Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2010843Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.2011152Z 2025-09-09T16:31:41.2011325Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2012350Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 166, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2013306Z y1 = F.silu(F.linear(cur_x, w1)) 2025-09-09T16:31:41.2013536Z 2025-09-09T16:31:41.2013707Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2014678Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 167, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2015580Z y3 = F.linear(cur_x, w3) 2025-09-09T16:31:41.2015777Z 2025-09-09T16:31:41.2015950Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2016942Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2017896Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.2018149Z 2025-09-09T16:31:41.2018382Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2019372Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 170, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2020321Z cur_out = F.linear(y1 * y3, y2) # [T'(e), D] 2025-09-09T16:31:41.2020576Z 2025-09-09T16:31:41.2020706Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2021077Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2021435Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2021799Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2022160Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2022512Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2022872Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2023279Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2024281Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 174, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2025260Z ordered_outs = torch.cat(outs, dim=0) # [T*A, D] 2025-09-09T16:31:41.2025543Z 2025-09-09T16:31:41.2025714Z cudagraph partition due to non gpu ops. Found from : 2025-09-09T16:31:41.2026690Z File "/opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/moe_quant/quantizable_moe_modules.py", line 179, in torch_dynamo_resume_in_forward_at_152 2025-09-09T16:31:41.2027648Z ordered_outs * ordered_token_activation_weights 2025-09-09T16:31:41.2027930Z 2025-09-09T16:31:41.2028057Z cudagraph partition due to non gpu ops 2025-09-09T16:31:41.2028624Z PASSED 2025-09-09T16:31:41.2029349Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_fake_dim_0_single_token PASSED 2025-09-09T16:31:41.2030528Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8wo_fake_dim_1_multiple_tokens PASSED 2025-09-09T16:31:41.2031557Z test/quantization/test_observer.py::TestQuantFlow::test_block_size_calc_success PASSED 2025-09-09T16:31:41.2032503Z test/quantization/test_observer.py::TestQuantFlow::test_block_size_row_errors PASSED 2025-09-09T16:31:41.2033436Z test/quantization/test_observer.py::TestQuantFlow::test_fixed_qparams_observer PASSED 2025-09-09T16:31:41.2034457Z test/quantization/test_observer.py::TestQuantFlow::test_min_max_per_channel_affine PASSED 2025-09-09T16:31:41.2035443Z test/quantization/test_observer.py::TestQuantFlow::test_min_max_per_tensor_affine PASSED 2025-09-09T16:31:41.2036437Z test/quantization/test_observer.py::TestQuantFlow::test_mse_observer PASSED 2025-09-09T16:31:41.2037484Z test/quantization/test_observer.py::TestLinearObserver::test_linear_observer_tensor_observe_weight_False PASSED 2025-09-09T16:31:41.2038743Z test/quantization/test_observer.py::TestLinearObserver::test_linear_observer_tensor_observe_weight_True PASSED 2025-09-09T16:31:41.2039797Z test/quantization/test_qat.py::TestQAT::test_composable_qat_quantizer PASSED 2025-09-09T16:31:41.2040683Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_dtype PASSED 2025-09-09T16:31:41.2041671Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_dynamic_and_range_learning PASSED 2025-09-09T16:31:41.2042603Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_eps PASSED 2025-09-09T16:31:41.2043489Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_granularity PASSED 2025-09-09T16:31:41.2044479Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_granularity_error_cases PASSED 2025-09-09T16:31:41.2045451Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_mapping_type PASSED 2025-09-09T16:31:41.2046378Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_config_torch_intx PASSED 2025-09-09T16:32:10.2574595Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_channel_group PASSED 2025-09-09T16:32:10.2575882Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token PASSED 2025-09-09T16:32:10.2576728Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token_vs_convert_bfloat16 PASSED 2025-09-09T16:32:10.2577676Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token_vs_convert_float16 PASSED 2025-09-09T16:32:10.2578799Z test/quantization/test_qat.py::TestQAT::test_fake_quantize_per_token_vs_convert_float32 PASSED 2025-09-09T16:32:10.2579606Z test/quantization/test_qat.py::TestQAT::test_fake_quantized_embedding_4w PASSED 2025-09-09T16:32:10.2580284Z test/quantization/test_qat.py::TestQAT::test_fake_quantized_linear_4w PASSED 2025-09-09T16:32:10.2580956Z test/quantization/test_qat.py::TestQAT::test_fake_quantized_linear_8da4w PASSED 2025-09-09T16:32:10.2581733Z test/quantization/test_qat.py::TestQAT::test_fake_quantizer_range_learning_is_symmetric_False PASSED 2025-09-09T16:32:10.2582592Z test/quantization/test_qat.py::TestQAT::test_fake_quantizer_range_learning_is_symmetric_True PASSED 2025-09-09T16:32:10.2583318Z test/quantization/test_qat.py::TestQAT::test_fake_quantizer_repr PASSED 2025-09-09T16:32:10.2583974Z test/quantization/test_qat.py::TestQAT::test_fbgemm_fp8_primitives SKIPPED 2025-09-09T16:32:10.2584685Z test/quantization/test_qat.py::TestQAT::test_fbgemm_int4_preshuffled_primitives SKIPPED 2025-09-09T16:32:10.2585402Z test/quantization/test_qat.py::TestQAT::test_float8_fake_quantize_config PASSED 2025-09-09T16:32:10.2586114Z test/quantization/test_qat.py::TestQAT::test_float8_fake_quantize_granularity0 PASSED 2025-09-09T16:32:10.2586841Z test/quantization/test_qat.py::TestQAT::test_float8_fake_quantize_granularity1 PASSED 2025-09-09T16:32:10.2587530Z test/quantization/test_qat.py::TestQAT::test_infer_fp8_int4_config PASSED 2025-09-09T16:32:10.2588203Z test/quantization/test_qat.py::TestQAT::test_infer_int4_weight_only_config PASSED 2025-09-09T16:32:10.2588888Z test/quantization/test_qat.py::TestQAT::test_legacy_quantize_api_e2e PASSED 2025-09-09T16:32:10.2589531Z test/quantization/test_qat.py::TestQAT::test_qat_4w_embedding PASSED 2025-09-09T16:32:10.2590192Z test/quantization/test_qat.py::TestQAT::test_qat_4w_linear tensor(5.9366e-05, device='cuda:0', dtype=torch.bfloat16, 2025-09-09T16:32:10.2590729Z grad_fn=) 2025-09-09T16:32:10.2591008Z PASSED 2025-09-09T16:32:10.2591482Z test/quantization/test_qat.py::TestQAT::test_qat_4w_primitives tensor(0.0005, device='cuda:0', dtype=torch.bfloat16) 2025-09-09T16:32:10.2592040Z PASSED 2025-09-09T16:32:10.2592588Z test/quantization/test_qat.py::TestQAT::test_qat_4w_quantizer tensor(0.0038, device='cuda:0', dtype=torch.bfloat16, grad_fn=) 2025-09-09T16:32:10.2593344Z PASSED 2025-09-09T16:32:10.2593870Z test/quantization/test_qat.py::TestQAT::test_qat_4w_quantizer_gradients PASSED 2025-09-09T16:32:10.2594510Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_eps PASSED 2025-09-09T16:32:10.2595180Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_linear PASSED 2025-09-09T16:32:10.2595876Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_prepare_vs_convert_bfloat16 PASSED 2025-09-09T16:32:10.2596684Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_prepare_vs_convert_float16 PASSED 2025-09-09T16:32:10.2597435Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_prepare_vs_convert_float32 PASSED 2025-09-09T16:32:10.2598119Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer PASSED 2025-09-09T16:32:10.2598812Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_disable_fake_quant PASSED 2025-09-09T16:32:10.2599622Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_disable_fake_quant_backward PASSED 2025-09-09T16:32:10.2600493Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_gradients PASSED 2025-09-09T16:32:10.2601207Z test/quantization/test_qat.py::TestQAT::test_qat_8da4w_quantizer_meta_weights PASSED 2025-09-09T16:32:10.2601927Z test/quantization/test_qat.py::TestQAT::test_qat_api_convert_no_quantization PASSED 2025-09-09T16:32:10.2602593Z test/quantization/test_qat.py::TestQAT::test_qat_api_deprecation PASSED 2025-09-09T16:32:10.2603211Z test/quantization/test_qat.py::TestQAT::test_qat_config_init PASSED 2025-09-09T16:32:10.2603827Z test/quantization/test_qat.py::TestQAT::test_qat_fp8a4w_quantizer PASSED 2025-09-09T16:32:10.2604446Z test/quantization/test_qat.py::TestQAT::test_qat_linear_bias PASSED 2025-09-09T16:32:10.2605123Z test/quantization/test_qat.py::TestQAT::test_qat_nvfp4_use_per_tensor_scale_False PASSED 2025-09-09T16:32:10.2605872Z test/quantization/test_qat.py::TestQAT::test_qat_nvfp4_use_per_tensor_scale_True PASSED 2025-09-09T16:32:10.2606547Z test/quantization/test_qat.py::TestQAT::test_qat_prototype_bc PASSED 2025-09-09T16:32:10.2607237Z test/quantization/test_qat.py::TestQAT::test_qat_range_learning_is_symmetric_False PASSED 2025-09-09T16:32:10.2607992Z test/quantization/test_qat.py::TestQAT::test_qat_range_learning_is_symmetric_True PASSED 2025-09-09T16:32:10.2608663Z test/quantization/test_qat.py::TestQAT::test_quantize_api_e2e PASSED 2025-09-09T16:32:10.2609285Z test/quantization/test_qat.py::TestQAT::test_quantize_api_errors PASSED 2025-09-09T16:32:10.2610029Z test/quantization/test_qat.py::TestQAT::test_quantize_api_fp8_fp8_granularity0 SKIPPED 2025-09-09T16:32:10.2610770Z test/quantization/test_qat.py::TestQAT::test_quantize_api_fp8_fp8_granularity1 SKIPPED 2025-09-09T16:32:10.2611466Z test/quantization/test_qat.py::TestQAT::test_quantize_api_fp8_int4 SKIPPED 2025-09-09T16:32:10.2612138Z test/quantization/test_qat.py::TestQAT::test_quantize_api_int4_version_1 SKIPPED 2025-09-09T16:32:10.2612834Z test/quantization/test_qat.py::TestQAT::test_quantize_api_int4_version_2 SKIPPED 2025-09-09T16:32:10.2613502Z test/quantization/test_qat.py::TestQAT::test_quantize_api_int8_int4 PASSED 2025-09-09T16:32:10.2614135Z test/quantization/test_qat.py::TestQAT::test_quantize_api_nvfp4 SKIPPED 2025-09-09T16:32:10.2614770Z test/quantization/test_qat.py::TestQAT::test_quantize_api_prepare PASSED 2025-09-09T16:32:10.2615400Z test/quantization/test_qat.py::TestQAT::test_replace_linear_8da4w PASSED 2025-09-09T16:32:10.2616031Z test/quantization/test_qat.py::TestQAT::test_replace_linear_int4 PASSED 2025-09-09T16:32:10.2616691Z test/quantization/test_quant_api.py::TestQuantFlow::test_8da4w_quantizer PASSED 2025-09-09T16:32:10.2617495Z test/quantization/test_quant_api.py::TestQuantFlow::test_8da4w_quantizer_linear_bias PASSED 2025-09-09T16:32:10.2618341Z test/quantization/test_quant_api.py::TestQuantFlow::test_dynamic_quant_gpu_singleline PASSED 2025-09-09T16:32:10.2619257Z test/quantization/test_quant_api.py::TestQuantFlow::test_dynamic_quant_gpu_unified_api_eager_mode_impl SKIPPED 2025-09-09T16:32:10.2620182Z test/quantization/test_quant_api.py::TestQuantFlow::test_dynamic_quant_gpu_unified_api_unified_impl SKIPPED 2025-09-09T16:32:10.2620968Z test/quantization/test_quant_api.py::TestQuantFlow::test_eval_wrapper SKIPPED 2025-09-09T16:32:10.2621686Z test/quantization/test_quant_api.py::TestQuantFlow::test_eval_wrapper_llama3 SKIPPED 2025-09-09T16:32:10.2622444Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4_wo_quant_save_load SKIPPED 2025-09-09T16:32:10.2623280Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_2_use_hqq_False PASSED 2025-09-09T16:32:10.2624175Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_2_use_hqq_True PASSED 2025-09-09T16:32:10.2625108Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_3_use_hqq_False PASSED 2025-09-09T16:32:10.2625992Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_bfloat16_x_dim_3_use_hqq_True PASSED 2025-09-09T16:32:10.2626872Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_2_use_hqq_False PASSED 2025-09-09T16:32:10.2627742Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_2_use_hqq_True PASSED 2025-09-09T16:32:10.2628620Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_3_use_hqq_False PASSED 2025-09-09T16:32:10.2629490Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float16_x_dim_3_use_hqq_True PASSED 2025-09-09T16:32:10.2630369Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_2_use_hqq_False PASSED 2025-09-09T16:32:10.2631244Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_2_use_hqq_True PASSED 2025-09-09T16:32:10.2632114Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_3_use_hqq_False PASSED 2025-09-09T16:32:10.2632985Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cpu_float32_x_dim_3_use_hqq_True PASSED 2025-09-09T16:32:10.2633805Z test/quantization/test_quant_api.py::TestQuantFlow::test_int4wo_cuda_serialization PASSED 2025-09-09T16:32:10.2634574Z test/quantization/test_quant_api.py::TestQuantFlow::test_int8_wo_quant_save_load PASSED 2025-09-09T16:32:10.2635376Z test/quantization/test_quant_api.py::TestQuantFlow::test_int8wo_quantized_model_to_device PASSED 2025-09-09T16:32:10.2636187Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_default PASSED 2025-09-09T16:32:10.2637091Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_embedding_linear PASSED 2025-09-09T16:32:10.2637936Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_module_name PASSED 2025-09-09T16:32:10.2638735Z test/quantization/test_quant_api.py::TestQuantFlow::test_module_fqn_to_config_skip PASSED 2025-09-09T16:35:49.9405176Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_model_streaming PASSED 2025-09-09T16:35:49.9406311Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_8da4w_mapping_type0 PASSED 2025-09-09T16:35:49.9407467Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_8da4w_mapping_type1 PASSED 2025-09-09T16:35:49.9408537Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_int4 PASSED 2025-09-09T16:35:49.9409851Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_int8_wo PASSED 2025-09-09T16:35:49.9411010Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_save_load PASSED 2025-09-09T16:35:49.9412235Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantized_tensor_subclass_save_load_map_location PASSED 2025-09-09T16:35:49.9413317Z test/quantization/test_quant_api.py::TestQuantFlow::test_quantizer_int4_weight_only SKIPPED 2025-09-09T16:35:49.9414298Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config0 PASSED 2025-09-09T16:35:49.9415289Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config1 PASSED 2025-09-09T16:35:49.9416289Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config10 PASSED 2025-09-09T16:35:49.9417295Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config2 PASSED 2025-09-09T16:35:49.9418288Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config3 PASSED 2025-09-09T16:35:49.9419389Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config4 PASSED 2025-09-09T16:35:49.9420386Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config5 PASSED 2025-09-09T16:35:49.9421374Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config6 PASSED 2025-09-09T16:35:49.9422365Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config7 PASSED 2025-09-09T16:35:49.9423355Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config8 PASSED 2025-09-09T16:35:49.9424338Z test/quantization/test_quant_api.py::TestQuantFlow::test_workflow_e2e_numerics_config9 PASSED 2025-09-09T16:35:49.9425410Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_group_sym PASSED 2025-09-09T16:35:49.9426619Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_group_sym_no_clipping_err PASSED 2025-09-09T16:35:49.9427831Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_tensor_asym PASSED 2025-09-09T16:35:49.9429005Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_tensor_asym_eps PASSED 2025-09-09T16:35:49.9430183Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_tensor_sym PASSED 2025-09-09T16:35:49.9431118Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_choose_qparams_token_asym PASSED 2025-09-09T16:35:49.9432006Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_fake_quantize_affine PASSED 2025-09-09T16:35:49.9432918Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_fake_quantize_affine_cachemask PASSED 2025-09-09T16:35:49.9433866Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_get_group_qparams_symmetric PASSED 2025-09-09T16:35:49.9434823Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_get_group_qparams_symmetric_memory PASSED 2025-09-09T16:35:49.9435843Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_get_groupwise_affine_qparams PASSED 2025-09-09T16:35:49.9436994Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_groupwise_affine_dequantize_tensor_from_qparams PASSED 2025-09-09T16:35:49.9438079Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_groupwise_affine_quantize_tensor_from_qparams PASSED 2025-09-09T16:35:49.9439092Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_maybe_expand_scale_to_tensor_shape PASSED 2025-09-09T16:35:49.9440149Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_activation_per_token_abs_max PASSED 2025-09-09T16:35:49.9441240Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_activation_per_token_abs_max_dtype PASSED 2025-09-09T16:35:49.9442578Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_activation_per_token_abs_max_zero_input PASSED 2025-09-09T16:35:49.9443606Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_channel_asym PASSED 2025-09-09T16:35:49.9444588Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_channel_asym_4d PASSED 2025-09-09T16:35:49.9445657Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_channel_asym_4d_multi_dim_reduction PASSED 2025-09-09T16:35:49.9446719Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_group_sym PASSED 2025-09-09T16:35:49.9447683Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_quantize_dequantize_tensor_asym PASSED 2025-09-09T16:35:49.9448589Z test/quantization/test_quant_primitives.py::TestQuantPrimitives::test_raises PASSED 2025-09-09T16:35:49.9449350Z test/sparsity/test_activation24.py::test_sparse24_sm90_sparsify_identity SKIPPED 2025-09-09T16:35:49.9450103Z test/sparsity/test_activation24.py::test_sparse24_sm90_sparsify_identity_scaled SKIPPED 2025-09-09T16:35:49.9450834Z test/sparsity/test_activation24.py::test_sparse24_sm90_sparsify_srelu SKIPPED 2025-09-09T16:35:49.9451572Z test/sparsity/test_activation24.py::test_srelu_fp8_semi_sparse_activation_linear SKIPPED 2025-09-09T16:35:49.9452324Z test/sparsity/test_activation24.py::test_sparse24_fp8_sm90_cutlass_gemm_eye SKIPPED 2025-09-09T16:35:49.9453105Z test/sparsity/test_activation24.py::test_sparse24_fp8_sm90_cutlass_gemm_random_tensor SKIPPED 2025-09-09T16:35:49.9454081Z test/sparsity/test_fast_sparse_training.py::TestRuntimeSemiStructuredSparsity::test_runtime_weight_sparsification SKIPPED 2025-09-09T16:35:49.9455216Z test/sparsity/test_fast_sparse_training.py::TestRuntimeSemiStructuredSparsity::test_runtime_weight_sparsification_compile SKIPPED 2025-09-09T16:35:49.9456154Z test/sparsity/test_marlin.py::SparseMarlin24::test_pack_unpack_equivalence PASSED 2025-09-09T16:35:49.9456928Z test/sparsity/test_marlin.py::SparseMarlin24::test_quant_sparse_marlin_layout_compile PASSED 2025-09-09T16:35:49.9457734Z test/sparsity/test_marlin.py::SparseMarlin24::test_quant_sparse_marlin_layout_eager PASSED 2025-09-09T16:35:49.9458491Z test/sparsity/test_sparse_api.py::TestSemiStructuredSparse::test_sparse SKIPPED 2025-09-09T16:35:49.9459286Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_quant_semi_sparse_compile_False SKIPPED 2025-09-09T16:35:49.9460145Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_sparse_marlin_compile_False PASSED 2025-09-09T16:35:49.9460981Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_sparse_marlin_compile_True PASSED 2025-09-09T16:35:49.9461857Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 PASSED 2025-09-09T16:35:49.9462785Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1024 PASSED 2025-09-09T16:35:49.9463689Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_True_input_shape_1 PASSED 2025-09-09T16:35:49.9464833Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_True_input_shape_1024 PASSED 2025-09-09T16:35:49.9465719Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_False PASSED 2025-09-09T16:35:49.9466660Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_True PASSED 2025-09-09T16:35:49.9467416Z test/sparsity/test_supermask.py::TestSupermask::test_from_linear PASSED 2025-09-09T16:35:49.9468260Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_25_blocksize_2 PASSED 2025-09-09T16:35:49.9469207Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_25_blocksize_4 PASSED 2025-09-09T16:35:49.9470089Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_25_blocksize_8 PASSED 2025-09-09T16:35:49.9470970Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_5_blocksize_2 PASSED 2025-09-09T16:35:49.9471849Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_5_blocksize_4 PASSED 2025-09-09T16:35:49.9472721Z test/sparsity/test_supermask.py::TestSupermask::test_supermask_sparsity_level_0_5_blocksize_8 PASSED 2025-09-09T16:35:49.9473515Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_2x4 PASSED 2025-09-09T16:35:49.9474336Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_unstructured PASSED 2025-09-09T16:35:49.9475059Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_prepare PASSED 2025-09-09T16:35:49.9475712Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_squash_mask PASSED 2025-09-09T16:35:49.9476504Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_two_layer_mlp_unstructured PASSED 2025-09-09T16:45:32.0920883Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_two_layer_mlp_unstructured_custom_config PASSED 2025-09-09T16:45:32.0922122Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_1_is_training_False PASSED 2025-09-09T16:45:32.0923394Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_1_is_training_True PASSED 2025-09-09T16:45:32.0924633Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_4_is_training_False PASSED 2025-09-09T16:45:32.0925871Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cpu_batch_size_4_is_training_True PASSED 2025-09-09T16:45:32.0927102Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_1_is_training_False PASSED 2025-09-09T16:45:32.0928330Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_1_is_training_True PASSED 2025-09-09T16:45:32.0929549Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_4_is_training_False PASSED 2025-09-09T16:45:32.0930781Z test/test_ao_models.py::TorchAOBasicTestCase::test_ao_inference_mode_device_cuda_batch_size_4_is_training_True PASSED 2025-09-09T16:45:32.0931911Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cpu_compile_False PASSED 2025-09-09T16:45:32.0932951Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cpu_compile_True PASSED 2025-09-09T16:45:32.0933998Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cuda_compile_False PASSED 2025-09-09T16:45:32.0935042Z test/test_low_bit_optim.py::TestQuantize::test_bf16_stochastic_round_device_cuda_compile_True PASSED 2025-09-09T16:45:32.0936055Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_compile_device_cpu PASSED 2025-09-09T16:45:32.0937065Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_compile_device_cuda PASSED 2025-09-09T16:45:32.0938084Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_correctness_device_cpu PASSED 2025-09-09T16:45:32.0939122Z test/test_low_bit_optim.py::TestQuantize::test_quantize_4bit_with_qmap_correctness_device_cuda PASSED 2025-09-09T16:45:32.0940417Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_compile_device_cpu PASSED 2025-09-09T16:45:32.0941521Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_compile_device_cuda PASSED 2025-09-09T16:45:32.0942631Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_correctness_device_cpu PASSED 2025-09-09T16:45:32.0943669Z test/test_low_bit_optim.py::TestQuantize::test_quantize_8bit_with_qmap_correctness_device_cuda PASSED 2025-09-09T16:45:32.0944679Z test/test_low_bit_optim.py::TestOptim::test_optim_4bit_correctness_optim_name_Adam4bit SKIPPED 2025-09-09T16:45:32.0945658Z test/test_low_bit_optim.py::TestOptim::test_optim_4bit_correctness_optim_name_AdamW4bit SKIPPED 2025-09-09T16:45:32.0946628Z test/test_low_bit_optim.py::TestOptim::test_optim_8bit_correctness_optim_name_Adam8bit SKIPPED 2025-09-09T16:45:32.0947625Z test/test_low_bit_optim.py::TestOptim::test_optim_8bit_correctness_optim_name_AdamW8bit SKIPPED 2025-09-09T16:45:32.0948389Z test/test_low_bit_optim.py::TestOptim::test_optim_bf16_stochastic_round_correctness PASSED 2025-09-09T16:45:32.0949365Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_correctness_offload_grad_False_grad_accum_1 PASSED 2025-09-09T16:45:32.0950281Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_correctness_offload_grad_False_grad_accum_2 PASSED 2025-09-09T16:45:32.0951179Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_correctness_offload_grad_True_grad_accum_1 PASSED 2025-09-09T16:45:32.0951946Z test/test_low_bit_optim.py::TestOptim::test_optim_cpu_offload_save_load PASSED 2025-09-09T16:45:32.0952750Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam4bit_device_cpu PASSED 2025-09-09T16:45:32.0953708Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam4bit_device_cuda PASSED 2025-09-09T16:45:32.0954782Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam8bit_device_cpu PASSED 2025-09-09T16:45:32.0955858Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_Adam8bit_device_cuda PASSED 2025-09-09T16:45:32.0957021Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_AdamFp8_device_cpu PASSED 2025-09-09T16:45:32.0958093Z test/test_low_bit_optim.py::TestOptim::test_optim_default_dtype_bf16_optim_name_AdamFp8_device_cuda SKIPPED 2025-09-09T16:45:32.0959145Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_bfloat16_device_cpu PASSED 2025-09-09T16:45:32.0960159Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_bfloat16_device_cuda PASSED 2025-09-09T16:45:32.0961166Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_float32_device_cpu PASSED 2025-09-09T16:45:32.0962188Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam4bit_float32_device_cuda PASSED 2025-09-09T16:45:32.0963222Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_bfloat16_device_cpu PASSED 2025-09-09T16:45:32.0964464Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_bfloat16_device_cuda PASSED 2025-09-09T16:45:32.0965508Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_float32_device_cpu PASSED 2025-09-09T16:45:32.0966531Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_Adam8bit_float32_device_cuda PASSED 2025-09-09T16:45:32.0967563Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_bfloat16_device_cpu PASSED 2025-09-09T16:45:32.0968585Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_bfloat16_device_cuda SKIPPED 2025-09-09T16:45:32.0969694Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_float32_device_cpu PASSED 2025-09-09T16:45:32.0970794Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamFp8_float32_device_cuda SKIPPED 2025-09-09T16:45:32.0971872Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_bfloat16_device_cpu PASSED 2025-09-09T16:45:32.0972903Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_bfloat16_device_cuda PASSED 2025-09-09T16:45:32.0973922Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_float32_device_cpu PASSED 2025-09-09T16:45:32.0974945Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW4bit_float32_device_cuda PASSED 2025-09-09T16:45:32.0975980Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_bfloat16_device_cpu PASSED 2025-09-09T16:45:32.0977006Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_bfloat16_device_cuda PASSED 2025-09-09T16:45:32.0978033Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_float32_device_cpu PASSED 2025-09-09T16:45:32.0979132Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamW8bit_float32_device_cuda PASSED 2025-09-09T16:45:32.0980166Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_bfloat16_device_cpu PASSED 2025-09-09T16:45:32.0981205Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_bfloat16_device_cuda SKIPPED 2025-09-09T16:45:32.0982243Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_float32_device_cpu PASSED 2025-09-09T16:45:32.0983278Z test/test_low_bit_optim.py::TestOptim::test_optim_smoke_optim_name_AdamWFp8_float32_device_cuda SKIPPED 2025-09-09T16:45:32.0984261Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam4bit_device_cpu PASSED 2025-09-09T16:45:32.0985219Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam4bit_device_cuda PASSED 2025-09-09T16:45:32.0986178Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam8bit_device_cpu PASSED 2025-09-09T16:45:32.0987128Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_Adam8bit_device_cuda PASSED 2025-09-09T16:45:32.0988094Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_AdamFp8_device_cpu PASSED 2025-09-09T16:45:32.0989067Z test/test_low_bit_optim.py::TestOptim::test_param_groups_optim_name_AdamFp8_device_cuda SKIPPED 2025-09-09T16:45:32.0990034Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape0_device_cpu PASSED 2025-09-09T16:45:32.0990998Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape0_device_cuda PASSED 2025-09-09T16:45:32.0991959Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape1_device_cpu PASSED 2025-09-09T16:45:32.0992979Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass0_shape1_device_cuda PASSED 2025-09-09T16:45:32.0993951Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape0_device_cpu PASSED 2025-09-09T16:45:32.0994909Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape0_device_cuda PASSED 2025-09-09T16:45:32.0995877Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape1_device_cpu PASSED 2025-09-09T16:47:30.0831244Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass1_shape1_device_cuda PASSED 2025-09-09T16:47:30.0832104Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape0_device_cpu PASSED 2025-09-09T16:47:30.0833133Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape0_device_cuda SKIPPED 2025-09-09T16:47:30.0834173Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape1_device_cpu PASSED 2025-09-09T16:47:30.0835189Z test/test_low_bit_optim.py::TestOptim::test_subclass_slice_subclass2_shape1_device_cuda SKIPPED 2025-09-09T16:47:30.0836250Z test/test_low_bit_optim.py::TestFSDP2::test_fsdp2 I0909 16:45:32.228400 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 104683 2025-09-09T16:47:30.0837452Z I0909 16:45:32.360835 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 104684 2025-09-09T16:47:30.0838013Z dist init r=0, world=2 2025-09-09T16:47:30.0838233Z dist init r=1, world=2 2025-09-09T16:47:30.0839096Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T16:47:30.0840109Z warnings.warn( # warn only once 2025-09-09T16:47:30.0841005Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T16:47:30.0842093Z warnings.warn( # warn only once 2025-09-09T16:47:30.0843038Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T16:47:30.0843942Z warnings.warn( # warn only once 2025-09-09T16:47:30.0844954Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:4818: UserWarning: No device id is provided via `init_process_group` or `barrier `. Using the current device set by the user. 2025-09-09T16:47:30.0845863Z warnings.warn( # warn only once 2025-09-09T16:47:30.0846185Z PASSED 2025-09-09T16:47:30.0846973Z test/test_low_bit_optim.py::TestFSDP2::test_uneven_shard I0909 16:47:09.252755 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 0 with pid 107475 2025-09-09T16:47:30.0848016Z I0909 16:47:09.383011 915 site-packages/torch/testing/_internal/common_distributed.py:741] Started process 1 with pid 107476 2025-09-09T16:47:30.0848589Z dist init r=0, world=2 2025-09-09T16:47:30.0848807Z dist init r=1, world=2 2025-09-09T16:47:30.0849155Z PASSED 2025-09-09T16:47:30.0849678Z test/test_model_architecture.py::TestModels::test_ln_linear_activation_model_0_cpu PASSED 2025-09-09T16:47:30.0850467Z test/test_model_architecture.py::TestModels::test_ln_linear_activation_model_1_cuda PASSED 2025-09-09T16:47:30.0851220Z test/test_model_architecture.py::TestModels::test_toy_linear_model_0_cpu PASSED 2025-09-09T16:47:30.0851988Z test/test_model_architecture.py::TestModels::test_toy_linear_model_1_cuda PASSED 2025-09-09T16:47:30.0852695Z test/test_model_architecture.py::TestModels::test_transformer_block_0_cpu PASSED 2025-09-09T16:47:30.0853404Z test/test_model_architecture.py::TestModels::test_transformer_block_1_cuda PASSED 2025-09-09T16:47:30.0854390Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_2_mbits_2_bfloat16 PASSED 2025-09-09T16:47:30.0855343Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_2_mbits_2_float16 PASSED 2025-09-09T16:47:30.0856381Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_3_mbits_2_bfloat16 PASSED 2025-09-09T16:47:30.0857333Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_1_OC_2048_IC_4096_splitK_5_ebits_3_mbits_2_float16 PASSED 2025-09-09T16:47:30.0858335Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_2_mbits_2_bfloat16 PASSED 2025-09-09T16:47:30.0859379Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_2_mbits_2_float16 PASSED 2025-09-09T16:47:30.0860367Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_3_mbits_2_bfloat16 PASSED 2025-09-09T16:47:30.0861466Z test/test_ops.py::TestOps::test_quant_llm_linear_correctness_BS_2_OC_8192_IC_8192_splitK_6_ebits_3_mbits_2_float16 PASSED 2025-09-09T16:47:30.0862256Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_2_mbits_2_bfloat16 PASSED 2025-09-09T16:47:30.0862959Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_2_mbits_2_float16 PASSED 2025-09-09T16:47:30.0863870Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_3_mbits_2_bfloat16 PASSED 2025-09-09T16:47:30.0864537Z test/test_ops.py::TestOps::test_quant_llm_linear_ebits_3_mbits_2_float16 PASSED 2025-09-09T16:47:30.0865466Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.0866561Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.0867753Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0868920Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.0869970Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.0871116Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0872174Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.0873345Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.0874396Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0875546Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.0876642Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.0877820Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0878871Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.0880003Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.0881112Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0882286Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.0883330Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.0884379Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0885655Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.0886815Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.0887883Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0889058Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.0890085Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.0891234Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.0892290Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.0893519Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1096149Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1097362Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1098405Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1099453Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1100614Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1101664Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1102742Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1103898Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1104942Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1106090Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1107142Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1108311Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1109367Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1110704Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1111800Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1113404Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1114549Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1115722Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1116846Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1118051Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1119089Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1120339Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_120_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1121462Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1122613Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1123655Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1124811Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1125846Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1127033Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1128077Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1129207Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1130257Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1131490Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1132533Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1133688Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1134734Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1135911Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1136971Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1138114Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1139286Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1140516Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1141579Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1142703Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1143748Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1144914Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1146037Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1147178Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_16_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1148234Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1149349Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:30.1150427Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1151481Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:30.1152603Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:30.1153638Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:30.1154673Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:30.1155828Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:32.0049586Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:32.0050857Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:32.0051923Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:32.0052963Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_18_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:32.0053985Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:32.0055011Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_float32 SKIPPED 2025-09-09T16:47:32.0056036Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:32.0057263Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:32.0058358Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_float32 SKIPPED 2025-09-09T16:47:32.0059373Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_100_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:32.0060399Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_bfloat16 SKIPPED 2025-09-09T16:47:32.0061416Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_float32 SKIPPED 2025-09-09T16:47:32.0062434Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_32_mask_dtype0 SKIPPED 2025-09-09T16:47:32.0063552Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_bfloat16 SKIPPED 2025-09-09T16:47:32.0064747Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_float32 SKIPPED 2025-09-09T16:47:32.0065756Z test/test_ops.py::TestOps::test_scaled_dot_product_int8_op_batch_size_56_n_head_2_q_seq_len_89_kv_seq_len_253_head_dim_64_mask_dtype0 SKIPPED 2025-09-09T16:47:32.0066667Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x4096-tiles_2] PASSED 2025-09-09T16:47:32.0067469Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x4096-tiles_4] PASSED 2025-09-09T16:47:32.0068277Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x4096-tiles_8] PASSED 2025-09-09T16:47:32.0069086Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x11008-tiles_2] PASSED 2025-09-09T16:47:32.0069886Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x11008-tiles_4] PASSED 2025-09-09T16:47:32.0070684Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x11008-tiles_8] PASSED 2025-09-09T16:47:32.0071476Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_11008x4096-tiles_2] PASSED 2025-09-09T16:47:32.0072283Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_11008x4096-tiles_4] PASSED 2025-09-09T16:47:32.0073087Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_11008x4096-tiles_8] PASSED 2025-09-09T16:47:32.0073888Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x14336-tiles_2] PASSED 2025-09-09T16:47:32.0074695Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x14336-tiles_4] PASSED 2025-09-09T16:47:32.0075490Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_4096x14336-tiles_8] PASSED 2025-09-09T16:47:32.0076363Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_14336x4096-tiles_2] PASSED 2025-09-09T16:47:32.0077158Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_14336x4096-tiles_4] PASSED 2025-09-09T16:47:32.0077950Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_correctness[shape_14336x4096-tiles_8] PASSED 2025-09-09T16:47:32.0078719Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x4096-tiles_2] PASSED 2025-09-09T16:47:32.0079440Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x4096-tiles_4] PASSED 2025-09-09T16:47:32.0080163Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x4096-tiles_8] PASSED 2025-09-09T16:47:32.0080968Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x11008-tiles_2] PASSED 2025-09-09T16:47:32.0081796Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x11008-tiles_4] PASSED 2025-09-09T16:47:32.0082577Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x11008-tiles_8] PASSED 2025-09-09T16:47:32.0083300Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_11008x4096-tiles_2] PASSED 2025-09-09T16:47:32.0084020Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_11008x4096-tiles_4] PASSED 2025-09-09T16:47:32.0084740Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_11008x4096-tiles_8] PASSED 2025-09-09T16:47:32.0085458Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x14336-tiles_2] PASSED 2025-09-09T16:47:32.0086184Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x14336-tiles_4] PASSED 2025-09-09T16:47:32.0086902Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_4096x14336-tiles_8] PASSED 2025-09-09T16:47:32.0087695Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_14336x4096-tiles_2] PASSED 2025-09-09T16:47:32.0088434Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_14336x4096-tiles_4] PASSED 2025-09-09T16:47:32.0089154Z test/test_ops.py::test_unpack_tensor_core_tiled_layout_op[shape_14336x4096-tiles_8] PASSED 2025-09-09T16:47:32.0089976Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-32] PASSED 2025-09-09T16:47:32.0090866Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-64] PASSED 2025-09-09T16:47:32.0091752Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-128] PASSED 2025-09-09T16:47:32.0092641Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-2-256] PASSED 2025-09-09T16:47:32.0093525Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-32] PASSED 2025-09-09T16:47:32.0094410Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-64] PASSED 2025-09-09T16:47:32.0095287Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-128] PASSED 2025-09-09T16:47:32.0096175Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-4-256] PASSED 2025-09-09T16:47:32.0097058Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-32] PASSED 2025-09-09T16:47:32.0097928Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-64] PASSED 2025-09-09T16:47:32.0098809Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-128] PASSED 2025-09-09T16:47:32.0099687Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 4096)-8-256] PASSED 2025-09-09T16:47:32.0100578Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-32] PASSED 2025-09-09T16:47:32.0101459Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-64] PASSED 2025-09-09T16:47:32.0102342Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-128] PASSED 2025-09-09T16:47:32.0103233Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-2-256] PASSED 2025-09-09T16:47:32.0104122Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-32] PASSED 2025-09-09T16:47:32.0105058Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-64] PASSED 2025-09-09T16:47:32.0105985Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-128] PASSED 2025-09-09T16:47:32.0106914Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-4-256] PASSED 2025-09-09T16:47:32.0107812Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-32] PASSED 2025-09-09T16:47:32.0108704Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-64] PASSED 2025-09-09T16:47:36.0753307Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-128] PASSED 2025-09-09T16:47:36.0754251Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 11008)-8-256] PASSED 2025-09-09T16:47:36.0755218Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-32] PASSED 2025-09-09T16:47:36.0756390Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-64] PASSED 2025-09-09T16:47:36.0757291Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-128] PASSED 2025-09-09T16:47:36.0758203Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-2-256] PASSED 2025-09-09T16:47:36.0759104Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-32] PASSED 2025-09-09T16:47:36.0759999Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-64] PASSED 2025-09-09T16:47:36.0760897Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-128] PASSED 2025-09-09T16:47:36.0761801Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-4-256] PASSED 2025-09-09T16:47:36.0762697Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-32] PASSED 2025-09-09T16:47:36.0763591Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-64] PASSED 2025-09-09T16:47:36.0764645Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-128] PASSED 2025-09-09T16:47:36.0765538Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(11008, 4096)-8-256] PASSED 2025-09-09T16:47:36.0766421Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-32] PASSED 2025-09-09T16:47:36.0767313Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-64] PASSED 2025-09-09T16:47:36.0768210Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-128] PASSED 2025-09-09T16:47:36.0769116Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-2-256] PASSED 2025-09-09T16:47:36.0770012Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-32] PASSED 2025-09-09T16:47:36.0770893Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-64] PASSED 2025-09-09T16:47:36.0771793Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-128] PASSED 2025-09-09T16:47:36.0772698Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-4-256] PASSED 2025-09-09T16:47:36.0773685Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-32] PASSED 2025-09-09T16:47:36.0774651Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-64] PASSED 2025-09-09T16:47:36.0775743Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-128] PASSED 2025-09-09T16:47:36.0776645Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(4096, 14336)-8-256] PASSED 2025-09-09T16:47:36.0777536Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-32] PASSED 2025-09-09T16:47:36.0778422Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-64] PASSED 2025-09-09T16:47:36.0779324Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-128] PASSED 2025-09-09T16:47:36.0780482Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-2-256] PASSED 2025-09-09T16:47:36.0781656Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-32] PASSED 2025-09-09T16:47:36.0782550Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-64] PASSED 2025-09-09T16:47:36.0783435Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-128] PASSED 2025-09-09T16:47:36.0784326Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-4-256] PASSED 2025-09-09T16:47:36.0785272Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-32] PASSED 2025-09-09T16:47:36.0786149Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-64] PASSED 2025-09-09T16:47:36.0787043Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-128] PASSED 2025-09-09T16:47:36.0787943Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_quant_dequant[(14336, 4096)-8-256] PASSED 2025-09-09T16:47:36.0788852Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-32] PASSED 2025-09-09T16:47:36.0789770Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-64] PASSED 2025-09-09T16:47:36.0790684Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-128] PASSED 2025-09-09T16:47:36.0791613Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-2-256] PASSED 2025-09-09T16:47:36.0792532Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-32] PASSED 2025-09-09T16:47:36.0793441Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-64] PASSED 2025-09-09T16:47:36.0794364Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-128] PASSED 2025-09-09T16:47:36.0795278Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-4-256] PASSED 2025-09-09T16:47:36.0796195Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-32] PASSED 2025-09-09T16:47:36.0797209Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-64] PASSED 2025-09-09T16:47:36.0798117Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-128] PASSED 2025-09-09T16:47:36.0799099Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 4096)-8-256] PASSED 2025-09-09T16:47:36.0800058Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-32] PASSED 2025-09-09T16:47:36.0801023Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-64] PASSED 2025-09-09T16:47:36.0801954Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-128] PASSED 2025-09-09T16:47:36.0802879Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-2-256] PASSED 2025-09-09T16:47:36.0803800Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-32] PASSED 2025-09-09T16:47:36.0804726Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-64] PASSED 2025-09-09T16:47:36.0805650Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-128] PASSED 2025-09-09T16:47:36.0806630Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-4-256] PASSED 2025-09-09T16:47:36.0807549Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-32] PASSED 2025-09-09T16:47:36.0808467Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-64] PASSED 2025-09-09T16:47:36.0809392Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-128] PASSED 2025-09-09T16:47:36.0810309Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 11008)-8-256] PASSED 2025-09-09T16:47:36.0811238Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-32] PASSED 2025-09-09T16:47:36.0812158Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-64] PASSED 2025-09-09T16:47:36.0813085Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-128] PASSED 2025-09-09T16:47:36.0814018Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-2-256] PASSED 2025-09-09T16:47:41.5001677Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-32] PASSED 2025-09-09T16:47:41.5002654Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-64] PASSED 2025-09-09T16:47:41.5003597Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-128] PASSED 2025-09-09T16:47:41.5004530Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-4-256] PASSED 2025-09-09T16:47:41.5005473Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-32] PASSED 2025-09-09T16:47:41.5006393Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-64] PASSED 2025-09-09T16:47:41.5007311Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-128] PASSED 2025-09-09T16:47:41.5008273Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(11008, 4096)-8-256] PASSED 2025-09-09T16:47:41.5009215Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-32] PASSED 2025-09-09T16:47:41.5010285Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-64] PASSED 2025-09-09T16:47:41.5011281Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-128] PASSED 2025-09-09T16:47:41.5012724Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-2-256] PASSED 2025-09-09T16:47:41.5013659Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-32] PASSED 2025-09-09T16:47:41.5014582Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-64] PASSED 2025-09-09T16:47:41.5015504Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-128] PASSED 2025-09-09T16:47:41.5016430Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-4-256] PASSED 2025-09-09T16:47:41.5017352Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-32] PASSED 2025-09-09T16:47:41.5018414Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-64] PASSED 2025-09-09T16:47:41.5019350Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-128] PASSED 2025-09-09T16:47:41.5020301Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(4096, 14336)-8-256] PASSED 2025-09-09T16:47:41.5021220Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-32] PASSED 2025-09-09T16:47:41.5022144Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-64] PASSED 2025-09-09T16:47:41.5023083Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-128] PASSED 2025-09-09T16:47:41.5024017Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-2-256] PASSED 2025-09-09T16:47:41.5033015Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-32] PASSED 2025-09-09T16:47:41.5033950Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-64] PASSED 2025-09-09T16:47:41.5034881Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-128] PASSED 2025-09-09T16:47:41.5035806Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-4-256] PASSED 2025-09-09T16:47:41.5036821Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-32] PASSED 2025-09-09T16:47:41.5037742Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-64] PASSED 2025-09-09T16:47:41.5038663Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-128] PASSED 2025-09-09T16:47:41.5039591Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_correctness_unpack_and_dequant[(14336, 4096)-8-256] PASSED 2025-09-09T16:47:41.5040405Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-32] PASSED 2025-09-09T16:47:41.5041107Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-64] PASSED 2025-09-09T16:47:41.5041815Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-128] PASSED 2025-09-09T16:47:41.5042520Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-2-256] PASSED 2025-09-09T16:47:41.5043317Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-32] PASSED 2025-09-09T16:47:41.5044016Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-64] PASSED 2025-09-09T16:47:41.5044768Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-128] PASSED 2025-09-09T16:47:41.5045522Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-4-256] PASSED 2025-09-09T16:47:41.5046222Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-32] PASSED 2025-09-09T16:47:41.5046929Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-64] PASSED 2025-09-09T16:47:41.5047626Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-128] PASSED 2025-09-09T16:47:41.5048331Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 4096)-8-256] PASSED 2025-09-09T16:47:41.5049090Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-32] PASSED 2025-09-09T16:47:41.5049796Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-64] PASSED 2025-09-09T16:47:41.5050566Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-128] PASSED 2025-09-09T16:47:41.5051278Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-2-256] PASSED 2025-09-09T16:47:41.5051994Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-32] PASSED 2025-09-09T16:47:41.5052704Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-64] PASSED 2025-09-09T16:47:41.5053406Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-128] PASSED 2025-09-09T16:47:41.5054126Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-4-256] PASSED 2025-09-09T16:47:41.5054828Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-32] PASSED 2025-09-09T16:47:41.5055536Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-64] PASSED 2025-09-09T16:47:41.5056247Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-128] PASSED 2025-09-09T16:47:41.5056963Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 11008)-8-256] PASSED 2025-09-09T16:47:41.5057677Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-32] PASSED 2025-09-09T16:47:41.5058435Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-64] PASSED 2025-09-09T16:47:41.5059140Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-128] PASSED 2025-09-09T16:47:41.5059854Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-2-256] PASSED 2025-09-09T16:47:41.5060560Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-32] PASSED 2025-09-09T16:47:41.5061258Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-64] PASSED 2025-09-09T16:47:41.5061960Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-128] PASSED 2025-09-09T16:47:41.5062673Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-4-256] PASSED 2025-09-09T16:47:41.5063382Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-32] PASSED 2025-09-09T16:47:41.5064341Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-64] PASSED 2025-09-09T16:47:41.5065054Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-128] PASSED 2025-09-09T16:47:41.5065765Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(11008, 4096)-8-256] PASSED 2025-09-09T16:47:41.5066479Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-32] PASSED 2025-09-09T16:47:41.5067285Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-64] PASSED 2025-09-09T16:47:41.5068055Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-128] PASSED 2025-09-09T16:47:41.5068853Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-2-256] PASSED 2025-09-09T16:47:41.5069562Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-32] PASSED 2025-09-09T16:47:41.5070268Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-64] PASSED 2025-09-09T16:47:45.0025696Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-128] PASSED 2025-09-09T16:47:45.0026480Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-4-256] PASSED 2025-09-09T16:47:45.0027198Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-32] PASSED 2025-09-09T16:47:45.0027936Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-64] PASSED 2025-09-09T16:47:45.0028826Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-128] PASSED 2025-09-09T16:47:45.0029555Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(4096, 14336)-8-256] PASSED 2025-09-09T16:47:45.0030285Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-32] PASSED 2025-09-09T16:47:45.0030997Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-64] PASSED 2025-09-09T16:47:45.0031723Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-128] PASSED 2025-09-09T16:47:45.0032451Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-2-256] PASSED 2025-09-09T16:47:45.0033162Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-32] PASSED 2025-09-09T16:47:45.0033876Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-64] PASSED 2025-09-09T16:47:45.0034589Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-128] PASSED 2025-09-09T16:47:45.0035317Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-4-256] PASSED 2025-09-09T16:47:45.0036032Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-32] PASSED 2025-09-09T16:47:45.0036818Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-64] PASSED 2025-09-09T16:47:45.0037529Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-128] PASSED 2025-09-09T16:47:45.0038247Z test/test_ops.py::test_dequantize_tensor_core_tiled_layout_op[(14336, 4096)-8-256] PASSED 2025-09-09T16:47:45.0038886Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0039421Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0039959Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0040501Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0041052Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0041601Z test/test_ops.py::test_marlin_24[1-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0042142Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0042683Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0043224Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0043771Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0044334Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0044966Z test/test_ops.py::test_marlin_24[1-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0045576Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0046176Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0046711Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0047254Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0047800Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0048352Z test/test_ops.py::test_marlin_24[1-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0048888Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0049425Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0049962Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0050517Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0051279Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0051832Z test/test_ops.py::test_marlin_24[1-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0052377Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0052905Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0053436Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0053979Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0054528Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0055080Z test/test_ops.py::test_marlin_24[4-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0055626Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0056171Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0056712Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0057266Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0057825Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0058380Z test/test_ops.py::test_marlin_24[4-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0058924Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0059460Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0059991Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0060525Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0061082Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0061638Z test/test_ops.py::test_marlin_24[4-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0062180Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0062718Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0063250Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0063978Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0064545Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0065093Z test/test_ops.py::test_marlin_24[4-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0065641Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0066244Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0066774Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0067379Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0067997Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0068551Z test/test_ops.py::test_marlin_24[8-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0069089Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0069628Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0070157Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0070706Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0071254Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0071814Z test/test_ops.py::test_marlin_24[8-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0072432Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0072957Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0073489Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0074025Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0074575Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0075125Z test/test_ops.py::test_marlin_24[8-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0075662Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0076208Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0076792Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0077348Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0077901Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0078458Z test/test_ops.py::test_marlin_24[8-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0079003Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0079531Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0080068Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0080613Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0081175Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T16:47:45.0081724Z test/test_ops.py::test_marlin_24[16-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T16:47:45.0082279Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T16:47:45.0082826Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T16:47:45.0083371Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T16:47:45.0083933Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T16:47:45.0084493Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7546638Z test/test_ops.py::test_marlin_24[16-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7547360Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7547907Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7548454Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7549167Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7549951Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7550605Z test/test_ops.py::test_marlin_24[16-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7551399Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7552025Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7552574Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7553166Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7553918Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7554491Z test/test_ops.py::test_marlin_24[16-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7555047Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7555694Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7556430Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7557126Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7557695Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7558412Z test/test_ops.py::test_marlin_24[32-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7559036Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7559585Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7560185Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7560898Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7561482Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7562059Z test/test_ops.py::test_marlin_24[32-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7562759Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7563411Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7564130Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7564681Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7565243Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7565806Z test/test_ops.py::test_marlin_24[32-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7566414Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7567093Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7567645Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7568209Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7568853Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7569513Z test/test_ops.py::test_marlin_24[32-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7570069Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7570605Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7571240Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7571906Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7572469Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7573031Z test/test_ops.py::test_marlin_24[64-128-512-4--1-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7573855Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7574535Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7575099Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7575834Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7576481Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7577052Z test/test_ops.py::test_marlin_24[64-128-512-4-128-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7577611Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7578254Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7578855Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7579402Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7579983Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7580811Z test/test_ops.py::test_marlin_24[64-128-512-8--1-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7581375Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(1, 1, 1)] PASSED 2025-09-09T16:47:46.7581921Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(1, 4, 8)] PASSED 2025-09-09T16:47:46.7582544Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(1, 7, 5)] PASSED 2025-09-09T16:47:46.7583223Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(13, 17, 67)] PASSED 2025-09-09T16:47:46.7583787Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(26, 37, 13)] PASSED 2025-09-09T16:47:46.7584357Z test/test_ops.py::test_marlin_24[64-128-512-8-128-(67, 13, 11)] PASSED 2025-09-09T16:47:46.7585093Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(1, 1, 1)] SKIPPED (...) 2025-09-09T16:47:46.7585716Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(1, 4, 8)] SKIPPED (...) 2025-09-09T16:47:46.7586309Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(1, 7, 5)] SKIPPED (...) 2025-09-09T16:47:46.7586899Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7587625Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7588293Z test/test_ops.py::test_marlin_qqq[1-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7588860Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.7589539Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.7590176Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.7590745Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7591319Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7592087Z test/test_ops.py::test_marlin_qqq[1-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7592675Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.7593239Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.7593841Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.7594559Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7595131Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7595696Z test/test_ops.py::test_marlin_qqq[1-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7596445Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.7597107Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.7597747Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.7598389Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7599164Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7599815Z test/test_ops.py::test_marlin_qqq[1-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7600373Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.7601049Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.7601690Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.7602245Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7602816Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7603585Z test/test_ops.py::test_marlin_qqq[1-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7604157Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.7604798Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.7605435Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.7606138Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7606710Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7607290Z test/test_ops.py::test_marlin_qqq[1-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7608055Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(1, 1, 1)] SKIPPED (...) 2025-09-09T16:47:46.7608654Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(1, 4, 8)] SKIPPED (...) 2025-09-09T16:47:46.7609242Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(1, 7, 5)] SKIPPED (...) 2025-09-09T16:47:46.7609858Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.7610591Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.7611154Z test/test_ops.py::test_marlin_qqq[4-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.7611719Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8033404Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8034222Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8034854Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8035437Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8036013Z test/test_ops.py::test_marlin_qqq[4-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8036670Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8037224Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8037777Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8038342Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8038910Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8039479Z test/test_ops.py::test_marlin_qqq[4-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8040041Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8040606Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8041168Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8041736Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8042438Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8043096Z test/test_ops.py::test_marlin_qqq[4-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8043666Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8044275Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8044833Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8045402Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8045970Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8046541Z test/test_ops.py::test_marlin_qqq[4-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8047103Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8047671Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8048236Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8048879Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8049461Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8050036Z test/test_ops.py::test_marlin_qqq[4-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8050614Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(1, 1, 1)] SKIPPED (...) 2025-09-09T16:47:46.8051197Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(1, 4, 8)] SKIPPED (...) 2025-09-09T16:47:46.8051834Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(1, 7, 5)] SKIPPED (...) 2025-09-09T16:47:46.8052414Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8052976Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8053547Z test/test_ops.py::test_marlin_qqq[8-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8054106Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8054670Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8055218Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8055788Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8056363Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8056931Z test/test_ops.py::test_marlin_qqq[8-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8057495Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8058042Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8058601Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8059169Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8059743Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8060320Z test/test_ops.py::test_marlin_qqq[8-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8060885Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8061448Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8062009Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8062588Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8063171Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8063913Z test/test_ops.py::test_marlin_qqq[8-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8064563Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8065180Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8065794Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8066368Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8066931Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8067502Z test/test_ops.py::test_marlin_qqq[8-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8068061Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8068624Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8069178Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8069754Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8070330Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8070976Z test/test_ops.py::test_marlin_qqq[8-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8071541Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8072086Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8072638Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8073197Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8073769Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8074336Z test/test_ops.py::test_marlin_qqq[16-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8074896Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8075461Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8076023Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8076671Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8077248Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8077819Z test/test_ops.py::test_marlin_qqq[16-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8078383Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8078942Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8079504Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8080072Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8080650Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8081224Z test/test_ops.py::test_marlin_qqq[16-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8081846Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8082429Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8082994Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8083573Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8084159Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8084755Z test/test_ops.py::test_marlin_qqq[16-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8085326Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8085882Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8086516Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8087133Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8087789Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8088375Z test/test_ops.py::test_marlin_qqq[16-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8088942Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8089509Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8090070Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8090653Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8091230Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8091947Z test/test_ops.py::test_marlin_qqq[16-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8092610Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8093165Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8504106Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8504939Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8505563Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8506143Z test/test_ops.py::test_marlin_qqq[32-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8506745Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8507527Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8508316Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8508888Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8509471Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8510053Z test/test_ops.py::test_marlin_qqq[32-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8510632Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8511189Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8511757Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8512331Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8512899Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8513470Z test/test_ops.py::test_marlin_qqq[32-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8514042Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8524002Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8524573Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8525144Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8525725Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8526296Z test/test_ops.py::test_marlin_qqq[32-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8526912Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8527715Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8528277Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8528988Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8529756Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8530552Z test/test_ops.py::test_marlin_qqq[32-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8531206Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8531998Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8532652Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8533229Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8533878Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8534677Z test/test_ops.py::test_marlin_qqq[32-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8535245Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8535814Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8536862Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8537428Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8537999Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8538791Z test/test_ops.py::test_marlin_qqq[64-128-64-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8539402Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8539961Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8540621Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8541348Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8541924Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8542497Z test/test_ops.py::test_marlin_qqq[64-128-64-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8543284Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8543902Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8544461Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8545188Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8545897Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8546463Z test/test_ops.py::test_marlin_qqq[64-128-128-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8547027Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8547814Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8548432Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8549013Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8549693Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8550433Z test/test_ops.py::test_marlin_qqq[64-128-128-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8551000Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8551562Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8552163Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8552721Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8553339Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8554203Z test/test_ops.py::test_marlin_qqq[64-128-256-4--1-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8554845Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(1, 1, 1)] SKIPPED 2025-09-09T16:47:46.8555450Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(1, 4, 8)] SKIPPED 2025-09-09T16:47:46.8556355Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(1, 7, 5)] SKIPPED 2025-09-09T16:47:46.8556960Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(13, 17, 67)] SKIPPED 2025-09-09T16:47:46.8557534Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(26, 37, 13)] SKIPPED 2025-09-09T16:47:46.8558305Z test/test_ops.py::test_marlin_qqq[64-128-256-4-128-(67, 13, 11)] SKIPPED 2025-09-09T16:47:46.8558950Z test/test_ops.py::test_swizzle_mm SKIPPED (ROCm not available) 2025-09-09T16:47:46.8559541Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8560314Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8561074Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8561841Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8562559Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8563393Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8564274Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8565021Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8565809Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8566438Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8567145Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8567917Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-2-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8568563Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8569276Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8570056Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8570711Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8571353Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8572100Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-128-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8572795Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8573437Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8574294Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8574960Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8575627Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8576353Z test/test_ops.py::test_scaled_embedding_bag_cpu[1-1024-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8577077Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8577715Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8578532Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8579189Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8579934Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8895150Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8896201Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8896848Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8897481Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8898120Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8898755Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8899392Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-2-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8900251Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8900949Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8901701Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8902440Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8903163Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8903807Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-128-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8904655Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8905321Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8905970Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8906656Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8907440Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8908110Z test/test_ops.py::test_scaled_embedding_bag_cpu[2-1024-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8908846Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8909598Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8910239Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8910870Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8911509Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8912137Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8912833Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8913553Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8914192Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8914901Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8915693Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8916444Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-2-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8917084Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8917798Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8918523Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8919287Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8920150Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8920874Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-128-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8921599Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8922329Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8923057Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8923723Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8924578Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8925257Z test/test_ops.py::test_scaled_embedding_bag_cpu[3-1024-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8925910Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8926564Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8927401Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8928047Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8928779Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8929539Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8930179Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8930815Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8931563Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8932240Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8932876Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8933769Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-2-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8934426Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8935060Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8935733Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8936506Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8937171Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8937910Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-128-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8938698Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-1-torch.int64] SKIPPED 2025-09-09T16:47:46.8939357Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-1-torch.int32] SKIPPED 2025-09-09T16:47:46.8940014Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-128-torch.int64] SKIPPED 2025-09-09T16:47:46.8940823Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-128-torch.int32] SKIPPED 2025-09-09T16:47:46.8941496Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-512-torch.int64] SKIPPED 2025-09-09T16:47:46.8942165Z test/test_ops.py::test_scaled_embedding_bag_cpu[10-1024-512-torch.int32] SKIPPED 2025-09-09T16:47:46.8943219Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype0-1-size_mnk0-False] SKIPPED 2025-09-09T16:47:46.8944206Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype1-1-size_mnk1-True] SKIPPED 2025-09-09T16:47:46.8945413Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype2-1-size_mnk2-False] SKIPPED 2025-09-09T16:47:46.8946457Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype3-1-size_mnk3-True] SKIPPED 2025-09-09T16:47:46.8947701Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype4-1-size_mnk4-False] SKIPPED 2025-09-09T16:47:46.8948697Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype5-1-size_mnk5-True] SKIPPED 2025-09-09T16:47:46.8949769Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype6-1-size_mnk6-False] SKIPPED 2025-09-09T16:47:46.8950797Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype7-1-size_mnk7-True] SKIPPED 2025-09-09T16:47:46.8951979Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype8-1-size_mnk8-False] SKIPPED 2025-09-09T16:47:46.8953042Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype9-1-size_mnk9-True] SKIPPED 2025-09-09T16:47:46.8954114Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype10-1-size_mnk10-False] SKIPPED 2025-09-09T16:47:46.8955192Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype11-1-size_mnk11-True] SKIPPED 2025-09-09T16:47:46.8956407Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype12-4-size_mnk12-False] SKIPPED 2025-09-09T16:47:46.8957469Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype13-4-size_mnk13-True] SKIPPED 2025-09-09T16:47:46.8958464Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype14-4-size_mnk14-False] SKIPPED 2025-09-09T16:47:46.8959616Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype15-4-size_mnk15-True] SKIPPED 2025-09-09T16:47:46.8960634Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype16-4-size_mnk16-False] SKIPPED 2025-09-09T16:47:46.9184283Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype17-4-size_mnk17-True] SKIPPED 2025-09-09T16:47:46.9185517Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype18-4-size_mnk18-False] SKIPPED 2025-09-09T16:47:46.9186522Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype19-4-size_mnk19-True] SKIPPED 2025-09-09T16:47:46.9187513Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype20-4-size_mnk20-False] SKIPPED 2025-09-09T16:47:46.9188518Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype21-4-size_mnk21-True] SKIPPED 2025-09-09T16:47:46.9189518Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype22-4-size_mnk22-False] SKIPPED 2025-09-09T16:47:46.9190520Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype23-4-size_mnk23-True] SKIPPED 2025-09-09T16:47:46.9191519Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype24-8-size_mnk24-False] SKIPPED 2025-09-09T16:47:46.9192510Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype25-8-size_mnk25-True] SKIPPED 2025-09-09T16:47:46.9193512Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype26-8-size_mnk26-False] SKIPPED 2025-09-09T16:47:46.9195372Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype27-8-size_mnk27-True] SKIPPED 2025-09-09T16:47:46.9196526Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype28-8-size_mnk28-False] SKIPPED 2025-09-09T16:47:46.9197588Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype29-8-size_mnk29-True] SKIPPED 2025-09-09T16:47:46.9198578Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype30-8-size_mnk30-False] SKIPPED 2025-09-09T16:47:46.9199575Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype31-8-size_mnk31-True] SKIPPED 2025-09-09T16:47:46.9200567Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype32-8-size_mnk32-False] SKIPPED 2025-09-09T16:47:46.9201552Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype33-8-size_mnk33-True] SKIPPED 2025-09-09T16:47:46.9202568Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype34-8-size_mnk34-False] SKIPPED 2025-09-09T16:47:46.9203629Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype35-8-size_mnk35-True] SKIPPED 2025-09-09T16:47:46.9204633Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype36-16-size_mnk36-False] SKIPPED 2025-09-09T16:47:46.9205632Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype37-16-size_mnk37-True] SKIPPED 2025-09-09T16:47:46.9206642Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype38-16-size_mnk38-False] SKIPPED 2025-09-09T16:47:46.9207642Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype39-16-size_mnk39-True] SKIPPED 2025-09-09T16:47:46.9208643Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype40-16-size_mnk40-False] SKIPPED 2025-09-09T16:47:46.9209650Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype41-16-size_mnk41-True] SKIPPED 2025-09-09T16:47:46.9210659Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype42-16-size_mnk42-False] SKIPPED 2025-09-09T16:47:46.9211653Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype43-16-size_mnk43-True] SKIPPED 2025-09-09T16:47:46.9212656Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype44-16-size_mnk44-False] SKIPPED 2025-09-09T16:47:46.9213647Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype45-16-size_mnk45-True] SKIPPED 2025-09-09T16:47:46.9214651Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype46-16-size_mnk46-False] SKIPPED 2025-09-09T16:47:46.9215658Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype47-16-size_mnk47-True] SKIPPED 2025-09-09T16:47:46.9216658Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype48-32-size_mnk48-False] SKIPPED 2025-09-09T16:47:46.9217663Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype49-32-size_mnk49-True] SKIPPED 2025-09-09T16:47:46.9218658Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype50-32-size_mnk50-False] SKIPPED 2025-09-09T16:47:46.9219665Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype51-32-size_mnk51-True] SKIPPED 2025-09-09T16:47:46.9220672Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype52-32-size_mnk52-False] SKIPPED 2025-09-09T16:47:46.9221802Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype53-32-size_mnk53-True] SKIPPED 2025-09-09T16:47:46.9222864Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype54-32-size_mnk54-False] SKIPPED 2025-09-09T16:47:46.9223869Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype55-32-size_mnk55-True] SKIPPED 2025-09-09T16:47:46.9224865Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype56-32-size_mnk56-False] SKIPPED 2025-09-09T16:47:46.9225867Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype57-32-size_mnk57-True] SKIPPED 2025-09-09T16:47:46.9226865Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype58-32-size_mnk58-False] SKIPPED 2025-09-09T16:47:46.9227865Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype59-32-size_mnk59-True] SKIPPED 2025-09-09T16:47:46.9228923Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype60-64-size_mnk60-False] SKIPPED 2025-09-09T16:47:46.9229918Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype61-64-size_mnk61-True] SKIPPED 2025-09-09T16:47:46.9230920Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype62-64-size_mnk62-False] SKIPPED 2025-09-09T16:47:46.9231920Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype63-64-size_mnk63-True] SKIPPED 2025-09-09T16:47:46.9232915Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype64-64-size_mnk64-False] SKIPPED 2025-09-09T16:47:46.9233915Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype65-64-size_mnk65-True] SKIPPED 2025-09-09T16:47:46.9234917Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype66-64-size_mnk66-False] SKIPPED 2025-09-09T16:47:46.9235922Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype67-64-size_mnk67-True] SKIPPED 2025-09-09T16:47:46.9237024Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype68-64-size_mnk68-False] SKIPPED 2025-09-09T16:47:46.9238015Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype69-64-size_mnk69-True] SKIPPED 2025-09-09T16:47:46.9239018Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype70-64-size_mnk70-False] SKIPPED 2025-09-09T16:47:46.9240016Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype71-64-size_mnk71-True] SKIPPED 2025-09-09T16:47:46.9241015Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype72-1-size_mnk72-False] SKIPPED 2025-09-09T16:47:46.9242013Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype73-1-size_mnk73-True] SKIPPED 2025-09-09T16:47:46.9243001Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype74-1-size_mnk74-False] SKIPPED 2025-09-09T16:47:46.9243991Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype75-1-size_mnk75-True] SKIPPED 2025-09-09T16:47:46.9244987Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype76-1-size_mnk76-False] SKIPPED 2025-09-09T16:47:46.9245970Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype77-1-size_mnk77-True] SKIPPED 2025-09-09T16:47:46.9467046Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype78-1-size_mnk78-False] SKIPPED 2025-09-09T16:47:46.9468470Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype79-1-size_mnk79-True] SKIPPED 2025-09-09T16:47:46.9469475Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype80-1-size_mnk80-False] SKIPPED 2025-09-09T16:47:46.9470471Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype81-1-size_mnk81-True] SKIPPED 2025-09-09T16:47:46.9471470Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype82-1-size_mnk82-False] SKIPPED 2025-09-09T16:47:46.9472503Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype83-1-size_mnk83-True] SKIPPED 2025-09-09T16:47:46.9473495Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype84-4-size_mnk84-False] SKIPPED 2025-09-09T16:47:46.9474550Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype85-4-size_mnk85-True] SKIPPED 2025-09-09T16:47:46.9475544Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype86-4-size_mnk86-False] SKIPPED 2025-09-09T16:47:46.9476667Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype87-4-size_mnk87-True] SKIPPED 2025-09-09T16:47:46.9477659Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype88-4-size_mnk88-False] SKIPPED 2025-09-09T16:47:46.9478653Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype89-4-size_mnk89-True] SKIPPED 2025-09-09T16:47:46.9479642Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype90-4-size_mnk90-False] SKIPPED 2025-09-09T16:47:46.9480642Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype91-4-size_mnk91-True] SKIPPED 2025-09-09T16:47:46.9481640Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype92-4-size_mnk92-False] SKIPPED 2025-09-09T16:47:46.9482625Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype93-4-size_mnk93-True] SKIPPED 2025-09-09T16:47:46.9483617Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype94-4-size_mnk94-False] SKIPPED 2025-09-09T16:47:46.9484611Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype95-4-size_mnk95-True] SKIPPED 2025-09-09T16:47:46.9485595Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype96-8-size_mnk96-False] SKIPPED 2025-09-09T16:47:46.9486591Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype97-8-size_mnk97-True] SKIPPED 2025-09-09T16:47:46.9487588Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype98-8-size_mnk98-False] SKIPPED 2025-09-09T16:47:46.9488582Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype99-8-size_mnk99-True] SKIPPED 2025-09-09T16:47:46.9489587Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype100-8-size_mnk100-False] SKIPPED 2025-09-09T16:47:46.9490589Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype101-8-size_mnk101-True] SKIPPED 2025-09-09T16:47:46.9491590Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype102-8-size_mnk102-False] SKIPPED 2025-09-09T16:47:46.9492732Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype103-8-size_mnk103-True] SKIPPED 2025-09-09T16:47:46.9493774Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype104-8-size_mnk104-False] SKIPPED 2025-09-09T16:47:46.9494820Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype105-8-size_mnk105-True] SKIPPED 2025-09-09T16:47:46.9495822Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype106-8-size_mnk106-False] SKIPPED 2025-09-09T16:47:46.9496828Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype107-8-size_mnk107-True] SKIPPED 2025-09-09T16:47:46.9497840Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype108-16-size_mnk108-False] SKIPPED 2025-09-09T16:47:46.9498987Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype109-16-size_mnk109-True] SKIPPED 2025-09-09T16:47:46.9500168Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype110-16-size_mnk110-False] SKIPPED 2025-09-09T16:47:46.9501345Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype111-16-size_mnk111-True] SKIPPED 2025-09-09T16:47:46.9502427Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype112-16-size_mnk112-False] SKIPPED 2025-09-09T16:47:46.9503490Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype113-16-size_mnk113-True] SKIPPED 2025-09-09T16:47:46.9504555Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype114-16-size_mnk114-False] SKIPPED 2025-09-09T16:47:46.9505587Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype115-16-size_mnk115-True] SKIPPED 2025-09-09T16:47:46.9506693Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype116-16-size_mnk116-False] SKIPPED 2025-09-09T16:47:46.9507721Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype117-16-size_mnk117-True] SKIPPED 2025-09-09T16:47:46.9508839Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype118-16-size_mnk118-False] SKIPPED 2025-09-09T16:47:46.9509847Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype119-16-size_mnk119-True] SKIPPED 2025-09-09T16:47:46.9510961Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype120-32-size_mnk120-False] SKIPPED 2025-09-09T16:47:46.9511982Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype121-32-size_mnk121-True] SKIPPED 2025-09-09T16:47:46.9513114Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype122-32-size_mnk122-False] SKIPPED 2025-09-09T16:47:46.9514152Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype123-32-size_mnk123-True] SKIPPED 2025-09-09T16:47:46.9515262Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype124-32-size_mnk124-False] SKIPPED 2025-09-09T16:47:46.9516341Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype125-32-size_mnk125-True] SKIPPED 2025-09-09T16:47:46.9517457Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype126-32-size_mnk126-False] SKIPPED 2025-09-09T16:47:46.9518473Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype127-32-size_mnk127-True] SKIPPED 2025-09-09T16:47:46.9519656Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype128-32-size_mnk128-False] SKIPPED 2025-09-09T16:47:46.9520737Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype129-32-size_mnk129-True] SKIPPED 2025-09-09T16:47:46.9521828Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype130-32-size_mnk130-False] SKIPPED 2025-09-09T16:47:46.9522913Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype131-32-size_mnk131-True] SKIPPED 2025-09-09T16:47:46.9523932Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype132-64-size_mnk132-False] SKIPPED 2025-09-09T16:47:46.9525028Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype133-64-size_mnk133-True] SKIPPED 2025-09-09T16:47:46.9526047Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype134-64-size_mnk134-False] SKIPPED 2025-09-09T16:47:46.9527231Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype135-64-size_mnk135-True] SKIPPED 2025-09-09T16:47:46.9528241Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype136-64-size_mnk136-False] SKIPPED 2025-09-09T16:47:46.9529347Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype137-64-size_mnk137-True] SKIPPED 2025-09-09T16:47:46.9530346Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype138-64-size_mnk138-False] SKIPPED 2025-09-09T16:47:46.9762232Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype139-64-size_mnk139-True] SKIPPED 2025-09-09T16:47:46.9763494Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype140-64-size_mnk140-False] SKIPPED 2025-09-09T16:47:46.9764700Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype141-64-size_mnk141-True] SKIPPED 2025-09-09T16:47:46.9765734Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype142-64-size_mnk142-False] SKIPPED 2025-09-09T16:47:46.9766754Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s4s4[dtype143-64-size_mnk143-True] SKIPPED 2025-09-09T16:47:46.9767753Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype0-1-size_mnk0-False] SKIPPED 2025-09-09T16:47:46.9768739Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype1-1-size_mnk1-True] SKIPPED 2025-09-09T16:47:46.9769718Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype2-1-size_mnk2-False] SKIPPED 2025-09-09T16:47:46.9770703Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype3-1-size_mnk3-True] SKIPPED 2025-09-09T16:47:46.9771686Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype4-1-size_mnk4-False] SKIPPED 2025-09-09T16:47:46.9772655Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype5-1-size_mnk5-True] SKIPPED 2025-09-09T16:47:46.9773635Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype6-1-size_mnk6-False] SKIPPED 2025-09-09T16:47:46.9774610Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype7-1-size_mnk7-True] SKIPPED 2025-09-09T16:47:46.9775582Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype8-1-size_mnk8-False] SKIPPED 2025-09-09T16:47:46.9776699Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype9-1-size_mnk9-True] SKIPPED 2025-09-09T16:47:46.9777748Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype10-1-size_mnk10-False] SKIPPED 2025-09-09T16:47:46.9778816Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype11-1-size_mnk11-True] SKIPPED 2025-09-09T16:47:46.9779821Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype12-4-size_mnk12-False] SKIPPED 2025-09-09T16:47:46.9780806Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype13-4-size_mnk13-True] SKIPPED 2025-09-09T16:47:46.9781837Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype14-4-size_mnk14-False] SKIPPED 2025-09-09T16:47:46.9782841Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype15-4-size_mnk15-True] SKIPPED 2025-09-09T16:47:46.9783847Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype16-4-size_mnk16-False] SKIPPED 2025-09-09T16:47:46.9784928Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype17-4-size_mnk17-True] SKIPPED 2025-09-09T16:47:46.9785914Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype18-4-size_mnk18-False] SKIPPED 2025-09-09T16:47:46.9786908Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype19-4-size_mnk19-True] SKIPPED 2025-09-09T16:47:46.9787909Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype20-4-size_mnk20-False] SKIPPED 2025-09-09T16:47:46.9788898Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype21-4-size_mnk21-True] SKIPPED 2025-09-09T16:47:46.9789894Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype22-4-size_mnk22-False] SKIPPED 2025-09-09T16:47:46.9790891Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype23-4-size_mnk23-True] SKIPPED 2025-09-09T16:47:46.9792234Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype24-8-size_mnk24-False] SKIPPED 2025-09-09T16:47:46.9793242Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype25-8-size_mnk25-True] SKIPPED 2025-09-09T16:47:46.9794362Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype26-8-size_mnk26-False] SKIPPED 2025-09-09T16:47:46.9795364Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype27-8-size_mnk27-True] SKIPPED 2025-09-09T16:47:46.9796449Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype28-8-size_mnk28-False] SKIPPED 2025-09-09T16:47:46.9797449Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype29-8-size_mnk29-True] SKIPPED 2025-09-09T16:47:46.9798564Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype30-8-size_mnk30-False] SKIPPED 2025-09-09T16:47:46.9799551Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype31-8-size_mnk31-True] SKIPPED 2025-09-09T16:47:46.9800653Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype32-8-size_mnk32-False] SKIPPED 2025-09-09T16:47:46.9801645Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype33-8-size_mnk33-True] SKIPPED 2025-09-09T16:47:46.9802722Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype34-8-size_mnk34-False] SKIPPED 2025-09-09T16:47:46.9803843Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype35-8-size_mnk35-True] SKIPPED 2025-09-09T16:47:46.9805028Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype36-16-size_mnk36-False] SKIPPED 2025-09-09T16:47:46.9806065Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype37-16-size_mnk37-True] SKIPPED 2025-09-09T16:47:46.9807175Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype38-16-size_mnk38-False] SKIPPED 2025-09-09T16:47:46.9808191Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype39-16-size_mnk39-True] SKIPPED 2025-09-09T16:47:46.9809323Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype40-16-size_mnk40-False] SKIPPED 2025-09-09T16:47:46.9810359Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype41-16-size_mnk41-True] SKIPPED 2025-09-09T16:47:46.9811529Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype42-16-size_mnk42-False] SKIPPED 2025-09-09T16:47:46.9812557Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype43-16-size_mnk43-True] SKIPPED 2025-09-09T16:47:46.9813661Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype44-16-size_mnk44-False] SKIPPED 2025-09-09T16:47:46.9814678Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype45-16-size_mnk45-True] SKIPPED 2025-09-09T16:47:46.9815685Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype46-16-size_mnk46-False] SKIPPED 2025-09-09T16:47:46.9816806Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype47-16-size_mnk47-True] SKIPPED 2025-09-09T16:47:46.9817821Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype48-32-size_mnk48-False] SKIPPED 2025-09-09T16:47:46.9818931Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype49-32-size_mnk49-True] SKIPPED 2025-09-09T16:47:46.9819941Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype50-32-size_mnk50-False] SKIPPED 2025-09-09T16:47:46.9821040Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype51-32-size_mnk51-True] SKIPPED 2025-09-09T16:47:46.9822098Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype52-32-size_mnk52-False] SKIPPED 2025-09-09T16:47:46.9823211Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype53-32-size_mnk53-True] SKIPPED 2025-09-09T16:47:46.9824227Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype54-32-size_mnk54-False] SKIPPED 2025-09-09T16:47:46.9825319Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype55-32-size_mnk55-True] SKIPPED 2025-09-09T16:47:47.0048043Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype56-32-size_mnk56-False] SKIPPED 2025-09-09T16:47:47.0049349Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype57-32-size_mnk57-True] SKIPPED 2025-09-09T16:47:47.0050361Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype58-32-size_mnk58-False] SKIPPED 2025-09-09T16:47:47.0051375Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype59-32-size_mnk59-True] SKIPPED 2025-09-09T16:47:47.0052514Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype60-64-size_mnk60-False] SKIPPED 2025-09-09T16:47:47.0053645Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype61-64-size_mnk61-True] SKIPPED 2025-09-09T16:47:47.0054669Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype62-64-size_mnk62-False] SKIPPED 2025-09-09T16:47:47.0055675Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype63-64-size_mnk63-True] SKIPPED 2025-09-09T16:47:47.0056687Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype64-64-size_mnk64-False] SKIPPED 2025-09-09T16:47:47.0057861Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype65-64-size_mnk65-True] SKIPPED 2025-09-09T16:47:47.0058871Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype66-64-size_mnk66-False] SKIPPED 2025-09-09T16:47:47.0060073Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype67-64-size_mnk67-True] SKIPPED 2025-09-09T16:47:47.0061082Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype68-64-size_mnk68-False] SKIPPED 2025-09-09T16:47:47.0062281Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype69-64-size_mnk69-True] SKIPPED 2025-09-09T16:47:47.0063294Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype70-64-size_mnk70-False] SKIPPED 2025-09-09T16:47:47.0064604Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype71-64-size_mnk71-True] SKIPPED 2025-09-09T16:47:47.0065610Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype72-1-size_mnk72-False] SKIPPED 2025-09-09T16:47:47.0066753Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype73-1-size_mnk73-True] SKIPPED 2025-09-09T16:47:47.0067772Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype74-1-size_mnk74-False] SKIPPED 2025-09-09T16:47:47.0068866Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype75-1-size_mnk75-True] SKIPPED 2025-09-09T16:47:47.0069871Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype76-1-size_mnk76-False] SKIPPED 2025-09-09T16:47:47.0070975Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype77-1-size_mnk77-True] SKIPPED 2025-09-09T16:47:47.0072042Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype78-1-size_mnk78-False] SKIPPED 2025-09-09T16:47:47.0073132Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype79-1-size_mnk79-True] SKIPPED 2025-09-09T16:47:47.0074160Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype80-1-size_mnk80-False] SKIPPED 2025-09-09T16:47:47.0075216Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype81-1-size_mnk81-True] SKIPPED 2025-09-09T16:47:47.0076338Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype82-1-size_mnk82-False] SKIPPED 2025-09-09T16:47:47.0077354Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype83-1-size_mnk83-True] SKIPPED 2025-09-09T16:47:47.0078458Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype84-4-size_mnk84-False] SKIPPED 2025-09-09T16:47:47.0079646Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype85-4-size_mnk85-True] SKIPPED 2025-09-09T16:47:47.0080887Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype86-4-size_mnk86-False] SKIPPED 2025-09-09T16:47:47.0081945Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype87-4-size_mnk87-True] SKIPPED 2025-09-09T16:47:47.0083111Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype88-4-size_mnk88-False] SKIPPED 2025-09-09T16:47:47.0084104Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype89-4-size_mnk89-True] SKIPPED 2025-09-09T16:47:47.0094489Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype90-4-size_mnk90-False] SKIPPED 2025-09-09T16:47:47.0095590Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype91-4-size_mnk91-True] SKIPPED 2025-09-09T16:47:47.0096724Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype92-4-size_mnk92-False] SKIPPED 2025-09-09T16:47:47.0097846Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype93-4-size_mnk93-True] SKIPPED 2025-09-09T16:47:47.0098987Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype94-4-size_mnk94-False] SKIPPED 2025-09-09T16:47:47.0099969Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype95-4-size_mnk95-True] SKIPPED 2025-09-09T16:47:47.0101040Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype96-8-size_mnk96-False] SKIPPED 2025-09-09T16:47:47.0102016Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype97-8-size_mnk97-True] SKIPPED 2025-09-09T16:47:47.0103098Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype98-8-size_mnk98-False] SKIPPED 2025-09-09T16:47:47.0104089Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype99-8-size_mnk99-True] SKIPPED 2025-09-09T16:47:47.0105180Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype100-8-size_mnk100-False] SKIPPED 2025-09-09T16:47:47.0106195Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype101-8-size_mnk101-True] SKIPPED 2025-09-09T16:47:47.0107194Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype102-8-size_mnk102-False] SKIPPED 2025-09-09T16:47:47.0108196Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype103-8-size_mnk103-True] SKIPPED 2025-09-09T16:47:47.0109418Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype104-8-size_mnk104-False] SKIPPED 2025-09-09T16:47:47.0110435Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype105-8-size_mnk105-True] SKIPPED 2025-09-09T16:47:47.0111545Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype106-8-size_mnk106-False] SKIPPED 2025-09-09T16:47:47.0112540Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype107-8-size_mnk107-True] SKIPPED 2025-09-09T16:47:47.0113677Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype108-16-size_mnk108-False] SKIPPED 2025-09-09T16:47:47.0114698Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype109-16-size_mnk109-True] SKIPPED 2025-09-09T16:47:47.0115875Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype110-16-size_mnk110-False] SKIPPED 2025-09-09T16:47:47.0117040Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype111-16-size_mnk111-True] SKIPPED 2025-09-09T16:47:47.0118226Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype112-16-size_mnk112-False] SKIPPED 2025-09-09T16:47:47.0119250Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype113-16-size_mnk113-True] SKIPPED 2025-09-09T16:47:47.0120350Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype114-16-size_mnk114-False] SKIPPED 2025-09-09T16:47:47.0121365Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype115-16-size_mnk115-True] SKIPPED 2025-09-09T16:47:47.0122474Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype116-16-size_mnk116-False] SKIPPED 2025-09-09T16:47:47.0309818Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype117-16-size_mnk117-True] SKIPPED 2025-09-09T16:47:47.0311108Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype118-16-size_mnk118-False] SKIPPED 2025-09-09T16:47:47.0312140Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype119-16-size_mnk119-True] SKIPPED 2025-09-09T16:47:47.0313226Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype120-32-size_mnk120-False] SKIPPED 2025-09-09T16:47:47.0314250Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype121-32-size_mnk121-True] SKIPPED 2025-09-09T16:47:47.0315339Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype122-32-size_mnk122-False] SKIPPED 2025-09-09T16:47:47.0316461Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype123-32-size_mnk123-True] SKIPPED 2025-09-09T16:47:47.0317496Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype124-32-size_mnk124-False] SKIPPED 2025-09-09T16:47:47.0318538Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype125-32-size_mnk125-True] SKIPPED 2025-09-09T16:47:47.0319605Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype126-32-size_mnk126-False] SKIPPED 2025-09-09T16:47:47.0320638Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype127-32-size_mnk127-True] SKIPPED 2025-09-09T16:47:47.0321724Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype128-32-size_mnk128-False] SKIPPED 2025-09-09T16:47:47.0322754Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype129-32-size_mnk129-True] SKIPPED 2025-09-09T16:47:47.0323846Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype130-32-size_mnk130-False] SKIPPED 2025-09-09T16:47:47.0324871Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype131-32-size_mnk131-True] SKIPPED 2025-09-09T16:47:47.0325958Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype132-64-size_mnk132-False] SKIPPED 2025-09-09T16:47:47.0326977Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype133-64-size_mnk133-True] SKIPPED 2025-09-09T16:47:47.0328061Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype134-64-size_mnk134-False] SKIPPED 2025-09-09T16:47:47.0329208Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype135-64-size_mnk135-True] SKIPPED 2025-09-09T16:47:47.0330368Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype136-64-size_mnk136-False] SKIPPED 2025-09-09T16:47:47.0331480Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype137-64-size_mnk137-True] SKIPPED 2025-09-09T16:47:47.0332574Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype138-64-size_mnk138-False] SKIPPED 2025-09-09T16:47:47.0333604Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype139-64-size_mnk139-True] SKIPPED 2025-09-09T16:47:47.0334635Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype140-64-size_mnk140-False] SKIPPED 2025-09-09T16:47:47.0335710Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype141-64-size_mnk141-True] SKIPPED 2025-09-09T16:47:47.0336744Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype142-64-size_mnk142-False] SKIPPED 2025-09-09T16:47:47.0337903Z test/test_ops_rowwise_scaled_linear_cutlass.py::test_rowwise_scaled_linear_cutlass_s8s4[dtype143-64-size_mnk143-True] SKIPPED 2025-09-09T16:47:47.0339022Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype0-Xq_Wq_dtypes0-1-size_mnk0-False] SKIPPED 2025-09-09T16:47:47.0340293Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype1-Xq_Wq_dtypes1-1-size_mnk1-True] SKIPPED 2025-09-09T16:47:47.0341495Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype2-Xq_Wq_dtypes2-1-size_mnk2-False] SKIPPED 2025-09-09T16:47:47.0342740Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype3-Xq_Wq_dtypes3-1-size_mnk3-True] SKIPPED 2025-09-09T16:47:47.0343996Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype4-Xq_Wq_dtypes4-1-size_mnk4-False] SKIPPED 2025-09-09T16:47:47.0345190Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype5-Xq_Wq_dtypes5-1-size_mnk5-True] SKIPPED 2025-09-09T16:47:47.0346441Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype6-Xq_Wq_dtypes6-1-size_mnk6-False] SKIPPED 2025-09-09T16:47:47.0347635Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype7-Xq_Wq_dtypes7-1-size_mnk7-True] SKIPPED 2025-09-09T16:47:47.0348874Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype8-Xq_Wq_dtypes8-1-size_mnk8-False] SKIPPED 2025-09-09T16:47:47.0350080Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype9-Xq_Wq_dtypes9-1-size_mnk9-True] SKIPPED 2025-09-09T16:47:47.0351348Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype10-Xq_Wq_dtypes10-1-size_mnk10-False] SKIPPED 2025-09-09T16:47:47.0352567Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype11-Xq_Wq_dtypes11-1-size_mnk11-True] SKIPPED 2025-09-09T16:47:47.0353918Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype12-Xq_Wq_dtypes12-4-size_mnk12-False] SKIPPED 2025-09-09T16:47:47.0355129Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype13-Xq_Wq_dtypes13-4-size_mnk13-True] SKIPPED 2025-09-09T16:47:47.0356562Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype14-Xq_Wq_dtypes14-4-size_mnk14-False] SKIPPED 2025-09-09T16:47:47.0357966Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype15-Xq_Wq_dtypes15-4-size_mnk15-True] SKIPPED 2025-09-09T16:47:47.0359187Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype16-Xq_Wq_dtypes16-4-size_mnk16-False] SKIPPED 2025-09-09T16:47:47.0360498Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype17-Xq_Wq_dtypes17-4-size_mnk17-True] SKIPPED 2025-09-09T16:47:47.0361722Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype18-Xq_Wq_dtypes18-4-size_mnk18-False] SKIPPED 2025-09-09T16:47:47.0363010Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype19-Xq_Wq_dtypes19-4-size_mnk19-True] SKIPPED 2025-09-09T16:47:47.0364378Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype20-Xq_Wq_dtypes20-4-size_mnk20-False] SKIPPED 2025-09-09T16:47:47.0365797Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype21-Xq_Wq_dtypes21-4-size_mnk21-True] SKIPPED 2025-09-09T16:47:47.0367017Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype22-Xq_Wq_dtypes22-4-size_mnk22-False] SKIPPED 2025-09-09T16:47:47.0368306Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype23-Xq_Wq_dtypes23-4-size_mnk23-True] SKIPPED 2025-09-09T16:47:47.0369524Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype24-Xq_Wq_dtypes24-1-size_mnk24-False] SKIPPED 2025-09-09T16:47:47.0370738Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype25-Xq_Wq_dtypes25-1-size_mnk25-True] SKIPPED 2025-09-09T16:47:47.0372052Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype26-Xq_Wq_dtypes26-1-size_mnk26-False] SKIPPED 2025-09-09T16:47:47.0373309Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype27-Xq_Wq_dtypes27-1-size_mnk27-True] SKIPPED 2025-09-09T16:47:47.0548837Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype28-Xq_Wq_dtypes28-1-size_mnk28-False] SKIPPED 2025-09-09T16:47:47.0550285Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype29-Xq_Wq_dtypes29-1-size_mnk29-True] SKIPPED 2025-09-09T16:47:47.0551511Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype30-Xq_Wq_dtypes30-1-size_mnk30-False] SKIPPED 2025-09-09T16:47:47.0552730Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype31-Xq_Wq_dtypes31-1-size_mnk31-True] SKIPPED 2025-09-09T16:47:47.0553952Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype32-Xq_Wq_dtypes32-1-size_mnk32-False] SKIPPED 2025-09-09T16:47:47.0555168Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype33-Xq_Wq_dtypes33-1-size_mnk33-True] SKIPPED 2025-09-09T16:47:47.0556479Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype34-Xq_Wq_dtypes34-1-size_mnk34-False] SKIPPED 2025-09-09T16:47:47.0557693Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype35-Xq_Wq_dtypes35-1-size_mnk35-True] SKIPPED 2025-09-09T16:47:47.0559107Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype36-Xq_Wq_dtypes36-4-size_mnk36-False] SKIPPED 2025-09-09T16:47:47.0560381Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype37-Xq_Wq_dtypes37-4-size_mnk37-True] SKIPPED 2025-09-09T16:47:47.0561596Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype38-Xq_Wq_dtypes38-4-size_mnk38-False] SKIPPED 2025-09-09T16:47:47.0562804Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype39-Xq_Wq_dtypes39-4-size_mnk39-True] SKIPPED 2025-09-09T16:47:47.0564167Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype40-Xq_Wq_dtypes40-4-size_mnk40-False] SKIPPED 2025-09-09T16:47:47.0565388Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype41-Xq_Wq_dtypes41-4-size_mnk41-True] SKIPPED 2025-09-09T16:47:47.0566707Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype42-Xq_Wq_dtypes42-4-size_mnk42-False] SKIPPED 2025-09-09T16:47:47.0567918Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype43-Xq_Wq_dtypes43-4-size_mnk43-True] SKIPPED 2025-09-09T16:47:47.0569134Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype44-Xq_Wq_dtypes44-4-size_mnk44-False] SKIPPED 2025-09-09T16:47:47.0570339Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype45-Xq_Wq_dtypes45-4-size_mnk45-True] SKIPPED 2025-09-09T16:47:47.0571553Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype46-Xq_Wq_dtypes46-4-size_mnk46-False] SKIPPED 2025-09-09T16:47:47.0572828Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype47-Xq_Wq_dtypes47-4-size_mnk47-True] SKIPPED 2025-09-09T16:47:47.0574039Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype48-Xq_Wq_dtypes48-1-size_mnk48-False] SKIPPED 2025-09-09T16:47:47.0575246Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype49-Xq_Wq_dtypes49-1-size_mnk49-True] SKIPPED 2025-09-09T16:47:47.0576445Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype50-Xq_Wq_dtypes50-1-size_mnk50-False] SKIPPED 2025-09-09T16:47:47.0577650Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype51-Xq_Wq_dtypes51-1-size_mnk51-True] SKIPPED 2025-09-09T16:47:47.0578864Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype52-Xq_Wq_dtypes52-1-size_mnk52-False] SKIPPED 2025-09-09T16:47:47.0580068Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype53-Xq_Wq_dtypes53-1-size_mnk53-True] SKIPPED 2025-09-09T16:47:47.0581271Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype54-Xq_Wq_dtypes54-1-size_mnk54-False] SKIPPED 2025-09-09T16:47:47.0582529Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype55-Xq_Wq_dtypes55-1-size_mnk55-True] SKIPPED 2025-09-09T16:47:47.0583725Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype56-Xq_Wq_dtypes56-1-size_mnk56-False] SKIPPED 2025-09-09T16:47:47.0585742Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype57-Xq_Wq_dtypes57-1-size_mnk57-True] SKIPPED 2025-09-09T16:47:47.0587070Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype58-Xq_Wq_dtypes58-1-size_mnk58-False] SKIPPED 2025-09-09T16:47:47.0588282Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype59-Xq_Wq_dtypes59-1-size_mnk59-True] SKIPPED 2025-09-09T16:47:47.0589492Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype60-Xq_Wq_dtypes60-4-size_mnk60-False] SKIPPED 2025-09-09T16:47:47.0590694Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype61-Xq_Wq_dtypes61-4-size_mnk61-True] SKIPPED 2025-09-09T16:47:47.0591900Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype62-Xq_Wq_dtypes62-4-size_mnk62-False] SKIPPED 2025-09-09T16:47:47.0593107Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype63-Xq_Wq_dtypes63-4-size_mnk63-True] SKIPPED 2025-09-09T16:47:47.0594361Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype64-Xq_Wq_dtypes64-4-size_mnk64-False] SKIPPED 2025-09-09T16:47:47.0595571Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype65-Xq_Wq_dtypes65-4-size_mnk65-True] SKIPPED 2025-09-09T16:47:47.0596875Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype66-Xq_Wq_dtypes66-4-size_mnk66-False] SKIPPED 2025-09-09T16:47:47.0598091Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype67-Xq_Wq_dtypes67-4-size_mnk67-True] SKIPPED 2025-09-09T16:47:47.0599316Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype68-Xq_Wq_dtypes68-4-size_mnk68-False] SKIPPED 2025-09-09T16:47:47.0600535Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype69-Xq_Wq_dtypes69-4-size_mnk69-True] SKIPPED 2025-09-09T16:47:47.0601739Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype70-Xq_Wq_dtypes70-4-size_mnk70-False] SKIPPED 2025-09-09T16:47:47.0603005Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype71-Xq_Wq_dtypes71-4-size_mnk71-True] SKIPPED 2025-09-09T16:47:47.0604218Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype72-Xq_Wq_dtypes72-1-size_mnk72-False] SKIPPED 2025-09-09T16:47:47.0605426Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype73-Xq_Wq_dtypes73-1-size_mnk73-True] SKIPPED 2025-09-09T16:47:47.0606645Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype74-Xq_Wq_dtypes74-1-size_mnk74-False] SKIPPED 2025-09-09T16:47:47.0607859Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype75-Xq_Wq_dtypes75-1-size_mnk75-True] SKIPPED 2025-09-09T16:47:47.0609071Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype76-Xq_Wq_dtypes76-1-size_mnk76-False] SKIPPED 2025-09-09T16:47:47.0610287Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype77-Xq_Wq_dtypes77-1-size_mnk77-True] SKIPPED 2025-09-09T16:47:47.0786679Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype78-Xq_Wq_dtypes78-1-size_mnk78-False] SKIPPED 2025-09-09T16:47:47.0788270Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype79-Xq_Wq_dtypes79-1-size_mnk79-True] SKIPPED 2025-09-09T16:47:47.0789543Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype80-Xq_Wq_dtypes80-1-size_mnk80-False] SKIPPED 2025-09-09T16:47:47.0790737Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype81-Xq_Wq_dtypes81-1-size_mnk81-True] SKIPPED 2025-09-09T16:47:47.0791986Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype82-Xq_Wq_dtypes82-1-size_mnk82-False] SKIPPED 2025-09-09T16:47:47.0793202Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype83-Xq_Wq_dtypes83-1-size_mnk83-True] SKIPPED 2025-09-09T16:47:47.0794400Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype84-Xq_Wq_dtypes84-4-size_mnk84-False] SKIPPED 2025-09-09T16:47:47.0795692Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype85-Xq_Wq_dtypes85-4-size_mnk85-True] SKIPPED 2025-09-09T16:47:47.0797006Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype86-Xq_Wq_dtypes86-4-size_mnk86-False] SKIPPED 2025-09-09T16:47:47.0798201Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype87-Xq_Wq_dtypes87-4-size_mnk87-True] SKIPPED 2025-09-09T16:47:47.0799403Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype88-Xq_Wq_dtypes88-4-size_mnk88-False] SKIPPED 2025-09-09T16:47:47.0800604Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype89-Xq_Wq_dtypes89-4-size_mnk89-True] SKIPPED 2025-09-09T16:47:47.0801805Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype90-Xq_Wq_dtypes90-4-size_mnk90-False] SKIPPED 2025-09-09T16:47:47.0803014Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype91-Xq_Wq_dtypes91-4-size_mnk91-True] SKIPPED 2025-09-09T16:47:47.0804216Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype92-Xq_Wq_dtypes92-4-size_mnk92-False] SKIPPED 2025-09-09T16:47:47.0805410Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype93-Xq_Wq_dtypes93-4-size_mnk93-True] SKIPPED 2025-09-09T16:47:47.0806604Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype94-Xq_Wq_dtypes94-4-size_mnk94-False] SKIPPED 2025-09-09T16:47:47.0807806Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype95-Xq_Wq_dtypes95-4-size_mnk95-True] SKIPPED 2025-09-09T16:47:47.0809011Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype96-Xq_Wq_dtypes96-1-size_mnk96-False] SKIPPED 2025-09-09T16:47:47.0810215Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype97-Xq_Wq_dtypes97-1-size_mnk97-True] SKIPPED 2025-09-09T16:47:47.0811410Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype98-Xq_Wq_dtypes98-1-size_mnk98-False] SKIPPED 2025-09-09T16:47:47.0812609Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype99-Xq_Wq_dtypes99-1-size_mnk99-True] SKIPPED 2025-09-09T16:47:47.0813882Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype100-Xq_Wq_dtypes100-1-size_mnk100-False] SKIPPED 2025-09-09T16:47:47.0815199Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype101-Xq_Wq_dtypes101-1-size_mnk101-True] SKIPPED 2025-09-09T16:47:47.0816432Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype102-Xq_Wq_dtypes102-1-size_mnk102-False] SKIPPED 2025-09-09T16:47:47.0817662Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype103-Xq_Wq_dtypes103-1-size_mnk103-True] SKIPPED 2025-09-09T16:47:47.0818885Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype104-Xq_Wq_dtypes104-1-size_mnk104-False] SKIPPED 2025-09-09T16:47:47.0820118Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype105-Xq_Wq_dtypes105-1-size_mnk105-True] SKIPPED 2025-09-09T16:47:47.0821402Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype106-Xq_Wq_dtypes106-1-size_mnk106-False] SKIPPED 2025-09-09T16:47:47.0822625Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype107-Xq_Wq_dtypes107-1-size_mnk107-True] SKIPPED 2025-09-09T16:47:47.0823855Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype108-Xq_Wq_dtypes108-4-size_mnk108-False] SKIPPED 2025-09-09T16:47:47.0825075Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype109-Xq_Wq_dtypes109-4-size_mnk109-True] SKIPPED 2025-09-09T16:47:47.0826304Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype110-Xq_Wq_dtypes110-4-size_mnk110-False] SKIPPED 2025-09-09T16:47:47.0827536Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype111-Xq_Wq_dtypes111-4-size_mnk111-True] SKIPPED 2025-09-09T16:47:47.0828762Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype112-Xq_Wq_dtypes112-4-size_mnk112-False] SKIPPED 2025-09-09T16:47:47.0829990Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype113-Xq_Wq_dtypes113-4-size_mnk113-True] SKIPPED 2025-09-09T16:47:47.0831219Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype114-Xq_Wq_dtypes114-4-size_mnk114-False] SKIPPED 2025-09-09T16:47:47.0832439Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype115-Xq_Wq_dtypes115-4-size_mnk115-True] SKIPPED 2025-09-09T16:47:47.0833671Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype116-Xq_Wq_dtypes116-4-size_mnk116-False] SKIPPED 2025-09-09T16:47:47.0834900Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype117-Xq_Wq_dtypes117-4-size_mnk117-True] SKIPPED 2025-09-09T16:47:47.0836121Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype118-Xq_Wq_dtypes118-4-size_mnk118-False] SKIPPED 2025-09-09T16:47:47.0837399Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype119-Xq_Wq_dtypes119-4-size_mnk119-True] SKIPPED 2025-09-09T16:47:47.0838624Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype120-Xq_Wq_dtypes120-1-size_mnk120-False] SKIPPED 2025-09-09T16:47:47.0839933Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype121-Xq_Wq_dtypes121-1-size_mnk121-True] SKIPPED 2025-09-09T16:47:47.0841205Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype122-Xq_Wq_dtypes122-1-size_mnk122-False] SKIPPED 2025-09-09T16:47:47.0842477Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype123-Xq_Wq_dtypes123-1-size_mnk123-True] SKIPPED 2025-09-09T16:47:47.0843708Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype124-Xq_Wq_dtypes124-1-size_mnk124-False] SKIPPED 2025-09-09T16:47:47.0844936Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype125-Xq_Wq_dtypes125-1-size_mnk125-True] SKIPPED 2025-09-09T16:47:47.0846159Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype126-Xq_Wq_dtypes126-1-size_mnk126-False] SKIPPED 2025-09-09T16:47:47.0847393Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype127-Xq_Wq_dtypes127-1-size_mnk127-True] SKIPPED 2025-09-09T16:47:47.1022084Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype128-Xq_Wq_dtypes128-1-size_mnk128-False] SKIPPED 2025-09-09T16:47:47.1023560Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype129-Xq_Wq_dtypes129-1-size_mnk129-True] SKIPPED 2025-09-09T16:47:47.1024788Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype130-Xq_Wq_dtypes130-1-size_mnk130-False] SKIPPED 2025-09-09T16:47:47.1026014Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype131-Xq_Wq_dtypes131-1-size_mnk131-True] SKIPPED 2025-09-09T16:47:47.1027242Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype132-Xq_Wq_dtypes132-4-size_mnk132-False] SKIPPED 2025-09-09T16:47:47.1028476Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype133-Xq_Wq_dtypes133-4-size_mnk133-True] SKIPPED 2025-09-09T16:47:47.1029699Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype134-Xq_Wq_dtypes134-4-size_mnk134-False] SKIPPED 2025-09-09T16:47:47.1030932Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype135-Xq_Wq_dtypes135-4-size_mnk135-True] SKIPPED 2025-09-09T16:47:47.1032158Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype136-Xq_Wq_dtypes136-4-size_mnk136-False] SKIPPED 2025-09-09T16:47:47.1033378Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype137-Xq_Wq_dtypes137-4-size_mnk137-True] SKIPPED 2025-09-09T16:47:47.1034612Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype138-Xq_Wq_dtypes138-4-size_mnk138-False] SKIPPED 2025-09-09T16:47:47.1035835Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype139-Xq_Wq_dtypes139-4-size_mnk139-True] SKIPPED 2025-09-09T16:47:47.1037154Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype140-Xq_Wq_dtypes140-4-size_mnk140-False] SKIPPED 2025-09-09T16:47:47.1038381Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype141-Xq_Wq_dtypes141-4-size_mnk141-True] SKIPPED 2025-09-09T16:47:47.1039605Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype142-Xq_Wq_dtypes142-4-size_mnk142-False] SKIPPED 2025-09-09T16:47:47.1041000Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype143-Xq_Wq_dtypes143-4-size_mnk143-True] SKIPPED 2025-09-09T16:47:47.1042285Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype144-Xq_Wq_dtypes144-1-size_mnk144-False] SKIPPED 2025-09-09T16:47:47.1043512Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype145-Xq_Wq_dtypes145-1-size_mnk145-True] SKIPPED 2025-09-09T16:47:47.1044730Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype146-Xq_Wq_dtypes146-1-size_mnk146-False] SKIPPED 2025-09-09T16:47:47.1045959Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype147-Xq_Wq_dtypes147-1-size_mnk147-True] SKIPPED 2025-09-09T16:47:47.1047187Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype148-Xq_Wq_dtypes148-1-size_mnk148-False] SKIPPED 2025-09-09T16:47:47.1048481Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype149-Xq_Wq_dtypes149-1-size_mnk149-True] SKIPPED 2025-09-09T16:47:47.1049709Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype150-Xq_Wq_dtypes150-1-size_mnk150-False] SKIPPED 2025-09-09T16:47:47.1050927Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype151-Xq_Wq_dtypes151-1-size_mnk151-True] SKIPPED 2025-09-09T16:47:47.1052204Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype152-Xq_Wq_dtypes152-1-size_mnk152-False] SKIPPED 2025-09-09T16:47:47.1053438Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype153-Xq_Wq_dtypes153-1-size_mnk153-True] SKIPPED 2025-09-09T16:47:47.1054665Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype154-Xq_Wq_dtypes154-1-size_mnk154-False] SKIPPED 2025-09-09T16:47:47.1055890Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype155-Xq_Wq_dtypes155-1-size_mnk155-True] SKIPPED 2025-09-09T16:47:47.1057115Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype156-Xq_Wq_dtypes156-4-size_mnk156-False] SKIPPED 2025-09-09T16:47:47.1058331Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype157-Xq_Wq_dtypes157-4-size_mnk157-True] SKIPPED 2025-09-09T16:47:47.1059557Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype158-Xq_Wq_dtypes158-4-size_mnk158-False] SKIPPED 2025-09-09T16:47:47.1060781Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype159-Xq_Wq_dtypes159-4-size_mnk159-True] SKIPPED 2025-09-09T16:47:47.1062007Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype160-Xq_Wq_dtypes160-4-size_mnk160-False] SKIPPED 2025-09-09T16:47:47.1063229Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype161-Xq_Wq_dtypes161-4-size_mnk161-True] SKIPPED 2025-09-09T16:47:47.1064613Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype162-Xq_Wq_dtypes162-4-size_mnk162-False] SKIPPED 2025-09-09T16:47:47.1065838Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype163-Xq_Wq_dtypes163-4-size_mnk163-True] SKIPPED 2025-09-09T16:47:47.1067203Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype164-Xq_Wq_dtypes164-4-size_mnk164-False] SKIPPED 2025-09-09T16:47:47.1068485Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype165-Xq_Wq_dtypes165-4-size_mnk165-True] SKIPPED 2025-09-09T16:47:47.1069717Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype166-Xq_Wq_dtypes166-4-size_mnk166-False] SKIPPED 2025-09-09T16:47:47.1070942Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype167-Xq_Wq_dtypes167-4-size_mnk167-True] SKIPPED 2025-09-09T16:47:47.1072159Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype168-Xq_Wq_dtypes168-1-size_mnk168-False] SKIPPED 2025-09-09T16:47:47.1073390Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype169-Xq_Wq_dtypes169-1-size_mnk169-True] SKIPPED 2025-09-09T16:47:47.1074684Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype170-Xq_Wq_dtypes170-1-size_mnk170-False] SKIPPED 2025-09-09T16:47:47.1075908Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype171-Xq_Wq_dtypes171-1-size_mnk171-True] SKIPPED 2025-09-09T16:47:47.1077244Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype172-Xq_Wq_dtypes172-1-size_mnk172-False] SKIPPED 2025-09-09T16:47:47.1078462Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype173-Xq_Wq_dtypes173-1-size_mnk173-True] SKIPPED 2025-09-09T16:47:47.1079692Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype174-Xq_Wq_dtypes174-1-size_mnk174-False] SKIPPED 2025-09-09T16:47:47.1080932Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype175-Xq_Wq_dtypes175-1-size_mnk175-True] SKIPPED 2025-09-09T16:47:47.1082161Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype176-Xq_Wq_dtypes176-1-size_mnk176-False] SKIPPED 2025-09-09T16:47:47.1346675Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype177-Xq_Wq_dtypes177-1-size_mnk177-True] SKIPPED 2025-09-09T16:47:47.1348213Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype178-Xq_Wq_dtypes178-1-size_mnk178-False] SKIPPED 2025-09-09T16:47:47.1349466Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype179-Xq_Wq_dtypes179-1-size_mnk179-True] SKIPPED 2025-09-09T16:47:47.1350706Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype180-Xq_Wq_dtypes180-4-size_mnk180-False] SKIPPED 2025-09-09T16:47:47.1351942Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype181-Xq_Wq_dtypes181-4-size_mnk181-True] SKIPPED 2025-09-09T16:47:47.1353509Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype182-Xq_Wq_dtypes182-4-size_mnk182-False] SKIPPED 2025-09-09T16:47:47.1355128Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype183-Xq_Wq_dtypes183-4-size_mnk183-True] SKIPPED 2025-09-09T16:47:47.1356878Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype184-Xq_Wq_dtypes184-4-size_mnk184-False] SKIPPED 2025-09-09T16:47:47.1358716Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype185-Xq_Wq_dtypes185-4-size_mnk185-True] SKIPPED 2025-09-09T16:47:47.1360360Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype186-Xq_Wq_dtypes186-4-size_mnk186-False] SKIPPED 2025-09-09T16:47:47.1362054Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype187-Xq_Wq_dtypes187-4-size_mnk187-True] SKIPPED 2025-09-09T16:47:47.1363931Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype188-Xq_Wq_dtypes188-4-size_mnk188-False] SKIPPED 2025-09-09T16:47:47.1365407Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype189-Xq_Wq_dtypes189-4-size_mnk189-True] SKIPPED 2025-09-09T16:47:47.1366648Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype190-Xq_Wq_dtypes190-4-size_mnk190-False] SKIPPED 2025-09-09T16:47:47.1368028Z test/test_ops_rowwise_scaled_linear_sparse_cutlass.py::test_rowwise_scaled_linear_sparse_cutlass_f8f8[dtype191-Xq_Wq_dtypes191-4-size_mnk191-True] SKIPPED 2025-09-09T16:47:47.1368948Z test/test_utils.py::TestTorchVersion::test_torch_version_at_least PASSED 2025-09-09T16:47:47.1369611Z test/test_utils.py::TestTorchVersion::test_torch_version_deprecation PASSED 2025-09-09T16:47:47.1370256Z test/test_utils.py::TestTorchAOBaseTensor::test_default_impls PASSED 2025-09-09T16:47:47.1370960Z test/test_utils.py::TestTorchAOBaseTensor::test_default_impls_with_optional_attr PASSED 2025-09-09T16:47:47.1371725Z test/test_utils.py::TestTorchAOBaseTensor::test_default_impls_with_optional_data PASSED 2025-09-09T16:47:47.1372432Z test/test_utils.py::TestTorchAOBaseTensor::test_print_arg_types PASSED 2025-09-09T16:47:47.1372872Z 2025-09-09T16:47:47.1373138Z =============================== warnings summary =============================== 2025-09-09T16:47:47.1373622Z ../../opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605 2025-09-09T16:47:47.1376099Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/__init__.py:1605: UserWarning: Please use the new API settings to control TF32 behavior, such as torch.backends.cudnn.conv.fp32_precision = 'tf32' or torch.backends.cuda.matmul.fp32_precision = 'ieee'. Old settings, e.g, torch.backends.cuda.matmul.allow_tf32 = True, torch.backends.cudnn.allow_tf32 = True, allowTF32CuDNN() and allowTF32CuBLAS() will be deprecated after Pytorch 2.9. Please see https://pytorch.org/docs/main/notes/cuda.html#tensorfloat-32-tf32-on-ampere-and-later-devices (Triggered internally at /pytorch/aten/src/ATen/Context.cpp:80.) 2025-09-09T16:47:47.1378494Z _C._set_float32_matmul_precision(precision) 2025-09-09T16:47:47.1378714Z 2025-09-09T16:47:47.1378944Z test/core/test_config.py::test_reconstructable_dict_file_round_trip[config8] 2025-09-09T16:47:47.1380081Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/core/config.py:250: UserWarning: Stored version is not the same as current default version of the config: stored_version=2, current_default_version=1, please check the deprecation warning 2025-09-09T16:47:47.1381076Z warnings.warn( 2025-09-09T16:47:47.1381203Z 2025-09-09T16:47:47.1381389Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_bfloat16 2025-09-09T16:47:47.1381827Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float16 2025-09-09T16:47:47.1382258Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float32 2025-09-09T16:47:47.1383502Z /pytorch/ao/test/dtypes/test_nf4.py:223: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T16:47:47.1384893Z torch.testing.assert_allclose(input_tensor, nf4_to_dtype, atol=0.13, rtol=0.13) 2025-09-09T16:47:47.1385231Z 2025-09-09T16:47:47.1385472Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_bfloat16 2025-09-09T16:47:47.1385964Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float16 2025-09-09T16:47:47.1386394Z test/dtypes/test_nf4.py::TestNF4Linear::test_to_copy_float32 2025-09-09T16:47:47.1387619Z /pytorch/ao/test/dtypes/test_nf4.py:229: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T16:47:47.1388779Z torch.testing.assert_allclose( 2025-09-09T16:47:47.1388962Z 2025-09-09T16:47:47.1389067Z test/float8/test_base.py: 36 warnings 2025-09-09T16:47:47.1390194Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/float8/float8_linear.py:261: DeprecationWarning: torch.get_autocast_gpu_dtype() is deprecated. Please use torch.get_autocast_dtype('cuda') instead. (Triggered internally at /pytorch/torch/csrc/autograd/init.cpp:856.) 2025-09-09T16:47:47.1391398Z autocast_dtype = torch.get_autocast_gpu_dtype() 2025-09-09T16:47:47.1391628Z 2025-09-09T16:47:47.1391848Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype3] 2025-09-09T16:47:47.1392356Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype4] 2025-09-09T16:47:47.1392861Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype5] 2025-09-09T16:47:47.1393356Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype6] 2025-09-09T16:47:47.1393857Z test/float8/test_float8_utils.py::test_non_float32_input[invalid_dtype7] 2025-09-09T16:47:47.1394857Z /pytorch/ao/test/float8/test_float8_utils.py:67: DeprecationWarning: an integer is required (got type float). Implicit conversion to integers using __int__ is deprecated, and may be removed in a future version of Python. 2025-09-09T16:47:47.1395818Z non_float32_tensor = torch.tensor([3.0], dtype=invalid_dtype) 2025-09-09T16:47:47.1396086Z 2025-09-09T16:47:47.1396495Z test/integration/test_integration.py::SmoothquantIntegrationTest::test_on_dummy_distilbert 2025-09-09T16:47:47.1397360Z /pytorch/ao/test/integration/test_integration.py:1440: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:47:47.1398027Z For migrations of users: 2025-09-09T16:47:47.1398732Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:47:47.1400057Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:47:47.1401222Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:47:47.1401859Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:47:47.1402285Z model_copy2 = torch.ao.quantization.quantize_dynamic( 2025-09-09T16:47:47.1402539Z 2025-09-09T16:47:47.1402839Z test/integration/test_integration.py::SmoothquantIntegrationTest::test_on_dummy_distilbert 2025-09-09T16:47:47.1403850Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/ao/quantization/quantize.py:566: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:47:47.1404657Z For migrations of users: 2025-09-09T16:47:47.1405352Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:47:47.1406677Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:47:47.1407932Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:47:47.1408604Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:47:47.1408984Z convert(model, mapping, inplace=True) 2025-09-09T16:47:47.1409182Z 2025-09-09T16:47:47.1409371Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_0_cuda 2025-09-09T16:47:47.1409881Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_mm_1_cuda 2025-09-09T16:47:47.1411156Z /pytorch/ao/test/kernel/test_autotuner.py:50: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T16:47:47.1412421Z torch.testing.assert_allclose(out32_1, out32_2) 2025-09-09T16:47:47.1412647Z 2025-09-09T16:47:47.1412865Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_0_cuda 2025-09-09T16:47:47.1413425Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_1_cpu 2025-09-09T16:47:47.1413935Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_2_cuda 2025-09-09T16:47:47.1414436Z test/kernel/test_autotuner.py::TestQuantFlow::test_int_scaled_mm_3_cpu 2025-09-09T16:47:47.1415722Z /pytorch/ao/test/kernel/test_autotuner.py:96: FutureWarning: `torch.testing.assert_allclose()` is deprecated since 1.12 and will be removed in a future release. Please use `torch.testing.assert_close()` instead. You can find detailed upgrade instructions in https://github.com/pytorch/pytorch/issues/61844. 2025-09-09T16:47:47.1416927Z torch.testing.assert_allclose(out32_1, out32_2) 2025-09-09T16:47:47.1417151Z 2025-09-09T16:47:47.1417450Z test/prototype/test_codebook_quant.py::TestCodebookQuantization::test_choose_qparams_codebook 2025-09-09T16:47:47.1418723Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/testing/_internal/common_utils.py:904: UserWarning: index_reduce() is in beta and the API may change at any time. (Triggered internally at /pytorch/aten/src/ATen/native/TensorAdvancedIndexing.cpp:1517.) 2025-09-09T16:47:47.1419779Z return callable(*args, **kwargs) 2025-09-09T16:47:47.1419966Z 2025-09-09T16:47:47.1420192Z test/prototype/test_parametrization.py::TestFakeSparsity::test_jit_trace 2025-09-09T16:47:47.1421740Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/sparsity/sparsifier/utils.py:134: TracerWarning: Converting a tensor to a Python boolean might cause the trace to be incorrect. We can't record the data flow of Python values, so this value will be treated as a constant in the future. This means that the trace might not generalize to other inputs! 2025-09-09T16:47:47.1423188Z assert self.mask.shape == x.shape 2025-09-09T16:47:47.1423384Z 2025-09-09T16:47:47.1423600Z test/prototype/test_scheduler.py::TestScheduler::test_lambda_scheduler 2025-09-09T16:47:47.1424104Z test/prototype/test_scheduler.py::TestCubicScheduler::test_step 2025-09-09T16:47:47.1425358Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/sparsity/scheduler/base_scheduler.py:133: UserWarning: Detected call of `scheduler.step()` before `sparsifier.step()`. You have to make sure you run the sparsifier.step() BEFORE any calls to the scheduler.step(). 2025-09-09T16:47:47.1426494Z warnings.warn( 2025-09-09T16:47:47.1426623Z 2025-09-09T16:47:47.1426942Z test/prototype/test_structured_sparsifier.py::TestBaseStructuredSparsifier::test_complex_conv2d 2025-09-09T16:47:47.1428103Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/prototype/sparsity/pruner/prune_functions.py:347: UserWarning: Converting a tensor with requires_grad=True to a scalar may lead to unexpected behavior. 2025-09-09T16:47:47.1429393Z Consider using tensor.detach() first. (Triggered internally at /pytorch/torch/csrc/autograd/generated/python_variable_methods.cpp:836.) 2025-09-09T16:47:47.1430093Z flattened_pruned_biases = torch.tensor( 2025-09-09T16:47:47.1430304Z 2025-09-09T16:47:47.1430605Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_conv_relu 2025-09-09T16:47:47.1431917Z /pytorch/ao/test/quantization/pt2e/test_graph_utils.py:42: FutureWarning: export(f, *args, **kwargs) is deprecated, use export(f)(*args, **kwargs) instead. If you don't migrate, we may break your export call in the future if your user defined kwargs conflict with future kwargs added to export(f). 2025-09-09T16:47:47.1433100Z m, guards = torchdynamo.export( # noqa: F841© 2025-09-09T16:47:47.1433335Z 2025-09-09T16:47:47.1433578Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_conv_bn_relu 2025-09-09T16:47:47.1434826Z /pytorch/ao/test/quantization/pt2e/test_graph_utils.py:86: FutureWarning: export(f, *args, **kwargs) is deprecated, use export(f)(*args, **kwargs) instead. If you don't migrate, we may break your export call in the future if your user defined kwargs conflict with future kwargs added to export(f). 2025-09-09T16:47:47.1436004Z m, guards = torchdynamo.export( # noqa: F841 2025-09-09T16:47:47.1436226Z 2025-09-09T16:47:47.1436594Z test/quantization/pt2e/test_graph_utils.py::TestGraphUtils::test_customized_equivalet_types_dict 2025-09-09T16:47:47.1437913Z /pytorch/ao/test/quantization/pt2e/test_graph_utils.py:118: FutureWarning: export(f, *args, **kwargs) is deprecated, use export(f)(*args, **kwargs) instead. If you don't migrate, we may break your export call in the future if your user defined kwargs conflict with future kwargs added to export(f). 2025-09-09T16:47:47.1439037Z m, guards = torchdynamo.export( # noqa: F841 2025-09-09T16:47:47.1439250Z 2025-09-09T16:47:47.1439418Z test/quantization/pt2e/test_quantize_pt2e.py: 18 warnings 2025-09-09T16:47:47.1439845Z test/quantization/pt2e/test_quantize_pt2e_qat.py: 91 warnings 2025-09-09T16:47:47.1440284Z test/quantization/pt2e/test_representation.py: 8 warnings 2025-09-09T16:47:47.1441046Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/testing/pt2e/_xnnpack_quantizer.py:289: UserWarning: XNNPACKQuantizer is deprecated! 2025-09-09T16:47:47.1441809Z warnings.warn(f"{self.__class__.__name__} is deprecated!") 2025-09-09T16:47:47.1442065Z 2025-09-09T16:47:47.1442394Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_allow_exported_model_train_eval 2025-09-09T16:47:47.1443078Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_disallow_eval_train 2025-09-09T16:47:47.1443812Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_annotate_mul_tensor 2025-09-09T16:47:47.1444620Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_annotate_mul_tensor 2025-09-09T16:47:47.1445432Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_annotate_mul_tensor 2025-09-09T16:47:47.1446248Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_conv2d_recipe 2025-09-09T16:47:47.1447063Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_linear_recipe 2025-09-09T16:47:47.1448186Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantize_pt2e.py:305: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:47:47.1449026Z For migrations of users: 2025-09-09T16:47:47.1449727Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:47:47.1451053Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:47:47.1452262Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:47:47.1452946Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:47:47.1453439Z return torch_convert_pt2e(model, use_reference_representation, fold_quantize) 2025-09-09T16:47:47.1453828Z 2025-09-09T16:47:47.1453998Z test/quantization/pt2e/test_quantize_pt2e.py: 384 warnings 2025-09-09T16:47:47.1454442Z test/quantization/pt2e/test_quantize_pt2e_qat.py: 300 warnings 2025-09-09T16:47:47.1455639Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/ao/quantization/pt2e/utils.py:359: FutureWarning: `torch.export.export_for_training` is deprecated and will be removed in PyTorch 2.10. Please use `torch.export.export` instead, which is functionally equivalent. 2025-09-09T16:47:47.1456778Z aten_pattern = torch.export.export_for_training( 2025-09-09T16:47:47.1457007Z 2025-09-09T16:47:47.1457354Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_embedding_conv_linear_quantization 2025-09-09T16:47:47.1458059Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_embedding_quantizer 2025-09-09T16:47:47.1459090Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/testing/pt2e/utils.py:108: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:47:47.1459878Z For migrations of users: 2025-09-09T16:47:47.1460573Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:47:47.1461897Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:47:47.1463050Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:47:47.1463688Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:47:47.1464251Z m_fx = prepare_fx( 2025-09-09T16:47:47.1464398Z 2025-09-09T16:47:47.1464670Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_model_is_exported 2025-09-09T16:47:47.1465950Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/fx/_symbolic_trace.py:922: UserWarning: Was not able to add assertion to guarantee correct input x to specialized function. It is up to the user to make sure that your inputs match the inputs you specialized the function with. 2025-09-09T16:47:47.1467015Z warnings.warn( 2025-09-09T16:47:47.1467149Z 2025-09-09T16:47:47.1467394Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_reentrant 2025-09-09T16:47:47.1468089Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_fold_bn_erases_bn_node 2025-09-09T16:47:47.1468872Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_fold_bn_erases_bn_node 2025-09-09T16:47:47.1469953Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/utils.py:145: UserWarning: must run observer before calling calculate_qparams. Returning default values. 2025-09-09T16:47:47.1470755Z warnings.warn( 2025-09-09T16:47:47.1470888Z 2025-09-09T16:47:47.1471134Z test/quantization/pt2e/test_quantize_pt2e.py::TestQuantizePT2E::test_reentrant 2025-09-09T16:47:47.1472221Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/observer.py:1350: UserWarning: must run observer before calling calculate_qparams. Returning default scale and zero point 2025-09-09T16:47:47.1473133Z warnings.warn( 2025-09-09T16:47:47.1473264Z 2025-09-09T16:47:47.1473644Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_bias_derived_qspec 2025-09-09T16:47:47.1474508Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_conv_bn_per_channel_weight_bias 2025-09-09T16:47:47.1475560Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn1d::test_qat_per_channel_weight_custom_dtype 2025-09-09T16:47:47.1476599Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_bias_derived_qspec 2025-09-09T16:47:47.1477461Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_conv_bn_per_channel_weight_bias 2025-09-09T16:47:47.1478341Z test/quantization/pt2e/test_quantize_pt2e_qat.py::TestQuantizePT2EQAT_ConvBn2d::test_qat_per_channel_weight_custom_dtype 2025-09-09T16:47:47.1479665Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/observer.py:253: UserWarning: Please use quant_min and quant_max to specify the range for observers. reduce_range will be deprecated in a future release of PyTorch. 2025-09-09T16:47:47.1480666Z warnings.warn( 2025-09-09T16:47:47.1480804Z 2025-09-09T16:47:47.1480981Z test/quantization/pt2e/test_quantize_pt2e_qat.py: 48 warnings 2025-09-09T16:47:47.1489245Z /pytorch/ao/test/quantization/pt2e/test_quantize_pt2e_qat.py:165: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:47:47.1490077Z For migrations of users: 2025-09-09T16:47:47.1490791Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:47:47.1492136Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:47:47.1493296Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:47:47.1493938Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:47:47.1494312Z model_fx = prepare_qat_fx( 2025-09-09T16:47:47.1494482Z 2025-09-09T16:47:47.1494663Z test/quantization/pt2e/test_quantize_pt2e_qat.py: 48 warnings 2025-09-09T16:47:47.1495608Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/ao/quantization/fx/prepare.py:464: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:47:47.1496449Z For migrations of users: 2025-09-09T16:47:47.1497163Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:47:47.1498523Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:47:47.1499701Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:47:47.1500352Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:47:47.1500854Z convert(root, mapping=module_to_qat_module, inplace=True, remove_qconfig=False) 2025-09-09T16:47:47.1501200Z 2025-09-09T16:47:47.1501502Z test/quantization/pt2e/test_x86inductor_fusion.py::TestPatternMatcher::test_qconv2d_add_3 2025-09-09T16:47:47.1502329Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_filter_conv2d_recipe 2025-09-09T16:47:47.1506826Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantizer/x86_inductor_quantizer.py:1325: UserWarning: The input of maxpool2d is not quantized, skip annotate maxpool2d with config QuantizationConfig(input_activation=QuantizationSpec(dtype=torch.uint8, observer_or_fake_quant_ctr=functools.partial(, eps=0.000244140625){}, quant_min=0, quant_max=255, qscheme=torch.per_tensor_affine, ch_axis=None, is_dynamic=False), output_activation=QuantizationSpec(dtype=torch.uint8, observer_or_fake_quant_ctr=functools.partial(, eps=0.000244140625){}, quant_min=0, quant_max=255, qscheme=torch.per_tensor_affine, ch_axis=None, is_dynamic=False), weight=QuantizationSpec(dtype=torch.int8, observer_or_fake_quant_ctr=functools.partial(, eps=0.000244140625){}, quant_min=-128, quant_max=127, qscheme=torch.per_channel_symmetric, ch_axis=0, is_dynamic=False), bias=None, is_qat=False). 2025-09-09T16:47:47.1511152Z warnings.warn( 2025-09-09T16:47:47.1511281Z 2025-09-09T16:47:47.1511627Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_q_attention_block 2025-09-09T16:47:47.1512453Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_q_attention_block 2025-09-09T16:47:47.1513298Z test/quantization/pt2e/test_x86inductor_fusion.py::TestDynamicPatternMatcher::test_qconv2d_maxpool2d_linear_dynamic_cpu 2025-09-09T16:47:47.1514804Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/_inductor/mkldnn_lowerings.py:736: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.detach().clone() or sourceTensor.detach().clone().requires_grad_(True), rather than torch.tensor(sourceTensor). 2025-09-09T16:47:47.1516108Z torch.tensor(w_zp_tensor, dtype=torch.int32), name=w_zp.get_name() 2025-09-09T16:47:47.1516545Z 2025-09-09T16:47:47.1517119Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_with_mixed_configs 2025-09-09T16:47:47.1518392Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantizer/x86_inductor_quantizer.py:484: UserWarning: Mixed dynamic and static quantization config is not supported. 2025-09-09T16:47:47.1519256Z warnings.warn( 2025-09-09T16:47:47.1519384Z 2025-09-09T16:47:47.1519856Z test/quantization/pt2e/test_x86inductor_quantizer.py::TestQuantizePT2EX86Inductor::test_set_module_name_and_module_type_with_mixed_configs 2025-09-09T16:47:47.1521170Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/pt2e/quantizer/x86_inductor_quantizer.py:383: UserWarning: Skip the quantization config for . 2025-09-09T16:47:47.1522086Z warnings.warn( 2025-09-09T16:47:47.1522219Z 2025-09-09T16:47:47.1522508Z test/quantization/test_moe_quant.py::TestMoEQuantCompile::test_int8dq_base_0_multiple_tokens 2025-09-09T16:47:47.1523209Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/_inductor/lowering.py:7184: UserWarning: 2025-09-09T16:47:47.1523796Z Online softmax is disabled on the fly since Inductor decides to 2025-09-09T16:47:47.1524233Z split the reduction. Cut an issue to PyTorch if this is an 2025-09-09T16:47:47.1524645Z important use case and you want to speed it up with online 2025-09-09T16:47:47.1524975Z softmax. 2025-09-09T16:47:47.1525152Z 2025-09-09T16:47:47.1525329Z warnings.warn( 2025-09-09T16:47:47.1525456Z 2025-09-09T16:47:47.1525661Z test/quantization/test_qat.py::TestQAT::test_legacy_quantize_api_e2e 2025-09-09T16:47:47.1526799Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/qat/utils.py:84: UserWarning: 'IntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: 2025-09-09T16:47:47.1527792Z 2025-09-09T16:47:47.1528096Z base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) 2025-09-09T16:47:47.1528564Z quantize_(model, QATConfig(base_config, step="prepare")) 2025-09-09T16:47:47.1528898Z # train (not shown) 2025-09-09T16:47:47.1529199Z quantize_(model, QATConfig(base_config, step="convert")) 2025-09-09T16:47:47.1529518Z 2025-09-09T16:47:47.1529801Z Alternatively, if you prefer to pass in fake quantization configs: 2025-09-09T16:47:47.1530156Z 2025-09-09T16:47:47.1530581Z activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) 2025-09-09T16:47:47.1531152Z weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) 2025-09-09T16:47:47.1531573Z qat_config = QATConfig( 2025-09-09T16:47:47.1531844Z activation_config=activation_config, 2025-09-09T16:47:47.1532175Z weight_config=weight_config, 2025-09-09T16:47:47.1532448Z step="prepare", 2025-09-09T16:47:47.1532669Z ) 2025-09-09T16:47:47.1532859Z quantize_(model, qat_config) 2025-09-09T16:47:47.1533100Z 2025-09-09T16:47:47.1533391Z Please see https://github.com/pytorch/ao/issues/2630 for more details. 2025-09-09T16:47:47.1533763Z 2025-09-09T16:47:47.1533944Z warnings.warn( 2025-09-09T16:47:47.1534077Z 2025-09-09T16:47:47.1534283Z test/quantization/test_qat.py::TestQAT::test_legacy_quantize_api_e2e 2025-09-09T16:47:47.1535437Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/quantization/qat/utils.py:84: UserWarning: 'FromIntXQuantizationAwareTrainingConfig' is deprecated and will be removed in a future release. Please use the following API instead: 2025-09-09T16:47:47.1536506Z 2025-09-09T16:47:47.1536812Z base_config = Int8DynamicActivationInt4WeightConfig(group_size=32) 2025-09-09T16:47:47.1537272Z quantize_(model, QATConfig(base_config, step="prepare")) 2025-09-09T16:47:47.1537609Z # train (not shown) 2025-09-09T16:47:47.1537902Z quantize_(model, QATConfig(base_config, step="convert")) 2025-09-09T16:47:47.1538217Z 2025-09-09T16:48:21.5578113Z Alternatively, if you prefer to pass in fake quantization configs: 2025-09-09T16:48:21.5582152Z 2025-09-09T16:48:21.5582602Z activation_config = IntxFakeQuantizeConfig(torch.int8, "per_token", is_symmetric=False) 2025-09-09T16:48:21.5583186Z weight_config = IntxFakeQuantizeConfig(torch.int4, group_size=32) 2025-09-09T16:48:21.5583596Z qat_config = QATConfig( 2025-09-09T16:48:21.5583876Z activation_config=activation_config, 2025-09-09T16:48:21.5584182Z weight_config=weight_config, 2025-09-09T16:48:21.5584459Z step="prepare", 2025-09-09T16:48:21.5584689Z ) 2025-09-09T16:48:21.5584884Z quantize_(model, qat_config) 2025-09-09T16:48:21.5585136Z 2025-09-09T16:48:21.5585447Z Please see https://github.com/pytorch/ao/issues/2630 for more details. 2025-09-09T16:48:21.5585822Z 2025-09-09T16:48:21.5586006Z warnings.warn( 2025-09-09T16:48:21.5586139Z 2025-09-09T16:48:21.5586356Z test/quantization/test_qat.py::TestQAT::test_qat_fp8a4w_quantizer 2025-09-09T16:48:21.5589436Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/autograd/graph.py:841: UserWarning: torchao::dequantize_affine_float8: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at /pytorch/torch/csrc/autograd/autograd_not_implemented_fallback.cpp:62.) 2025-09-09T16:48:21.5592587Z return Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass 2025-09-09T16:48:21.5592985Z 2025-09-09T16:48:21.5593246Z test/sparsity/test_marlin.py::SparseMarlin24::test_quant_sparse_marlin_layout_compile 2025-09-09T16:48:21.5593872Z test/sparsity/test_sparse_api.py::TestQuantSemiSparse::test_sparse_marlin_compile_True 2025-09-09T16:48:21.5597316Z /opt/conda/envs/venv/lib/python3.9/site-packages/torch/autograd/graph.py:841: UserWarning: torchao::marlin_24_gemm: an autograd kernel was not registered to the Autograd key(s) but we are trying to backprop through it. This may lead to silently incorrect behavior. This behavior is deprecated and will be removed in a future version of PyTorch. If your operator is differentiable, please ensure you have registered an autograd kernel to the correct Autograd key (e.g. DispatchKey::Autograd, DispatchKey::CompositeImplicitAutograd). If your operator is not differentiable, or to squash this warning and use the previous behavior, please register torch::CppFunction::makeFallthrough() to DispatchKey::Autograd. (Triggered internally at /pytorch/torch/csrc/autograd/autograd_not_implemented_fallback.cpp:62.) 2025-09-09T16:48:21.5600484Z return Variable._execution_engine.run_backward( # Calls into the C++ engine to run the backward pass 2025-09-09T16:48:21.5600884Z 2025-09-09T16:48:21.5601197Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 2025-09-09T16:48:21.5602848Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/sparsity/blocksparse.py:198: UserWarning: Sparse BSR tensor support is in beta state. If you miss a functionality in the sparse tensor support, please submit a feature request to https://github.com/pytorch/pytorch/issues. (Triggered internally at /pytorch/aten/src/ATen/SparseCsrTensorImpl.cpp:53.) 2025-09-09T16:48:21.5604392Z bsr_tensor = dense_tensor.to_sparse_bsr(blocksize) 2025-09-09T16:48:21.5604626Z 2025-09-09T16:48:21.5604937Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 2025-09-09T16:48:21.5606364Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=2048 K=1024 N=1 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T16:48:21.5607545Z warn_once( 2025-09-09T16:48:21.5607665Z 2025-09-09T16:48:21.5607977Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1 2025-09-09T16:48:21.5609405Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=1024 K=2048 N=1 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T16:48:21.5610586Z warn_once( 2025-09-09T16:48:21.5610706Z 2025-09-09T16:48:21.5611027Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1024 2025-09-09T16:48:21.5612470Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=2048 K=1024 N=1024 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T16:48:21.5613661Z warn_once( 2025-09-09T16:48:21.5613779Z 2025-09-09T16:48:21.5614100Z test/sparsity/test_sparse_api.py::TestBlockSparseWeight::test_sparse_compile_False_input_shape_1024 2025-09-09T16:48:21.5615545Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=1024 K=2048 N=1024 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.float16 out_dtype=torch.float16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T16:48:21.5616734Z warn_once( 2025-09-09T16:48:21.5616850Z 2025-09-09T16:48:21.5617123Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_False 2025-09-09T16:48:21.5618494Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=256 K=128 N=256 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.int8 out_dtype=torch.bfloat16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T16:48:21.5619711Z warn_once( 2025-09-09T16:48:21.5619827Z 2025-09-09T16:48:21.5620098Z test/sparsity/test_sparse_api.py::TestQuantBlockSparseWeight::test_sparse_compile_False 2025-09-09T16:48:21.5621540Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/kernel/bsr_triton_ops.py:240: UserWarning: bsr_dense_addmm uses non-optimal triton kernel parameters for M=128 K=256 N=256 Ms=64, Ks=64 beta=0 alpha=1 dtype=torch.int8 out_dtype=torch.bfloat16. To find optimal triton kernel parameters, run with BSR_AUTOTUNE=1 2025-09-09T16:48:21.5622705Z warn_once( 2025-09-09T16:48:21.5622834Z 2025-09-09T16:48:21.5623050Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_2x4 2025-09-09T16:48:21.5623993Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/sparsity/wanda.py:46: UserWarning: WandaSparsifier got semi_structured_bock_size=4, sparsity_level fixed to 50% (2:4) sparsity 2025-09-09T16:48:21.5624797Z warnings.warn( 2025-09-09T16:48:21.5624931Z 2025-09-09T16:48:21.5625151Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_2x4 2025-09-09T16:48:21.5625701Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_one_layer_mlp_unstructured 2025-09-09T16:48:21.5626284Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_prepare 2025-09-09T16:48:21.5626767Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_squash_mask 2025-09-09T16:48:21.5627301Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_two_layer_mlp_unstructured 2025-09-09T16:48:21.5627944Z test/sparsity/test_wanda.py::TestWandaSparsifier::test_two_layer_mlp_unstructured_custom_config 2025-09-09T16:48:21.5628917Z /opt/conda/envs/venv/lib/python3.9/site-packages/torchao/sparsity/wanda.py:75: DeprecationWarning: torch.ao.quantization is deprecated and will be removed in 2.10. 2025-09-09T16:48:21.5629691Z For migrations of users: 2025-09-09T16:48:21.5630390Z 1. Eager mode quantization (torch.ao.quantization.quantize, torch.ao.quantization.quantize_dynamic), please migrate to use torchao eager mode quantize_ API instead 2025-09-09T16:48:21.5631716Z 2. FX graph mode quantization (torch.ao.quantization.quantize_fx.prepare_fx,torch.ao.quantization.quantize_fx.convert_fx, please migrate to use torchao pt2e quantization API instead (prepare_pt2e, convert_pt2e) 2025-09-09T16:48:21.5632948Z 3. pt2e quantization has been migrated to torchao (https://github.com/pytorch/ao/tree/main/torchao/quantization/pt2e) 2025-09-09T16:48:21.5633597Z see https://github.com/pytorch/ao/issues/2259 for more details 2025-09-09T16:48:21.5634019Z torch.ao.quantization.prepare(model, inplace=True) 2025-09-09T16:48:21.5634264Z 2025-09-09T16:48:21.5634479Z -- Docs: https://docs.pytest.org/en/stable/how-to/capture-warnings.html 2025-09-09T16:48:21.5635509Z == 2681 passed, 4429 skipped, 28 xfailed, 1009 warnings in 8915.35s (2:28:35) == 2025-09-09T16:48:21.5679320Z ##[group]Run pmeier/pytest-results-action@a2c1430e2bddadbad9f49a6f9b879f062c6b19b1 2025-09-09T16:48:21.5679781Z with: 2025-09-09T16:48:21.5680052Z path: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:21.5680410Z fail-on-empty: false 2025-09-09T16:48:21.5680624Z env: 2025-09-09T16:48:21.5680854Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:21.5681177Z REPOSITORY: pytorch/ao 2025-09-09T16:48:21.5681406Z PR_NUMBER: 2963 2025-09-09T16:48:21.5682703Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:21.5684159Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:21.5684710Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:21.5685293Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:21.5685711Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:21.5686022Z ##[endgroup] 2025-09-09T16:48:21.6290487Z Prepare all required actions 2025-09-09T16:48:21.6329410Z ##[group]Run ./test-infra/.github/actions/chown-directory 2025-09-09T16:48:21.6329732Z with: 2025-09-09T16:48:21.6329992Z directory: /home/ec2-user/actions-runner/_work/ao/ao/ 2025-09-09T16:48:21.6330435Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T16:48:21.6330818Z env: 2025-09-09T16:48:21.6331054Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:21.6331381Z REPOSITORY: pytorch/ao 2025-09-09T16:48:21.6331617Z PR_NUMBER: 2963 2025-09-09T16:48:21.6332902Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:21.6334511Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:21.6335038Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:21.6346971Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:21.6347417Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:21.6347738Z ##[endgroup] 2025-09-09T16:48:21.6373988Z ##[group]Run docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T16:48:21.6374607Z docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T16:48:21.6389859Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T16:48:21.6390192Z env: 2025-09-09T16:48:21.6390427Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:21.6390759Z REPOSITORY: pytorch/ao 2025-09-09T16:48:21.6390988Z PR_NUMBER: 2963 2025-09-09T16:48:21.6392287Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:21.6393734Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:21.6394256Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:21.6394750Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:21.6395163Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:21.6395594Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T16:48:21.6396041Z DIRECTORY: /home/ec2-user/actions-runner/_work/ao/ao/ 2025-09-09T16:48:21.6396472Z ##[endgroup] 2025-09-09T16:48:21.6652757Z Unable to find image '308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest' locally 2025-09-09T16:48:21.8697175Z latest: Pulling from tool/alpine 2025-09-09T16:48:21.8697495Z 540db60ca938: Pulling fs layer 2025-09-09T16:48:21.9643401Z 540db60ca938: Download complete 2025-09-09T16:48:22.0785199Z 540db60ca938: Pull complete 2025-09-09T16:48:22.0894040Z Digest: sha256:def822f9851ca422481ec6fee59a9966f12b351c62ccb9aca841526ffaa9f748 2025-09-09T16:48:22.0935796Z Status: Downloaded newer image for 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest 2025-09-09T16:48:23.2264031Z Prepare all required actions 2025-09-09T16:48:23.2290630Z ##[group]Run ./test-infra/.github/actions/chown-directory 2025-09-09T16:48:23.2290953Z with: 2025-09-09T16:48:23.2291212Z directory: /home/ec2-user/actions-runner/_work/_temp 2025-09-09T16:48:23.2291786Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T16:48:23.2292160Z env: 2025-09-09T16:48:23.2292398Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:23.2292728Z REPOSITORY: pytorch/ao 2025-09-09T16:48:23.2292972Z PR_NUMBER: 2963 2025-09-09T16:48:23.2294268Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:23.2295760Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:23.2296310Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:23.2296802Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:23.2297226Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:23.2297546Z ##[endgroup] 2025-09-09T16:48:23.2323266Z ##[group]Run docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T16:48:23.2323989Z docker run --rm -v "${DIRECTORY}":/v -w /v "${ALPINE_IMAGE}" chown -R "$(id -u):$(id -g)" . 2025-09-09T16:48:23.2339052Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T16:48:23.2339385Z env: 2025-09-09T16:48:23.2339615Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:23.2339939Z REPOSITORY: pytorch/ao 2025-09-09T16:48:23.2340176Z PR_NUMBER: 2963 2025-09-09T16:48:23.2341456Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:23.2342905Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:23.2343431Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:23.2343919Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:23.2344332Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:23.2344774Z ALPINE_IMAGE: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine 2025-09-09T16:48:23.2345220Z DIRECTORY: /home/ec2-user/actions-runner/_work/_temp 2025-09-09T16:48:23.2345530Z ##[endgroup] 2025-09-09T16:48:24.5470181Z ##[group]Run # Only do these steps if we actually want to upload an artifact 2025-09-09T16:48:24.5470725Z # Only do these steps if we actually want to upload an artifact 2025-09-09T16:48:24.5471126Z if [[ -n "${UPLOAD_ARTIFACT_NAME}" ]]; then 2025-09-09T16:48:24.5471625Z  # If the default execution path is followed then we should get a wheel in the dist/ folder 2025-09-09T16:48:24.5472175Z  # attempt to just grab whatever is in there and scoop it all up 2025-09-09T16:48:24.5472684Z  if find "dist/" -name "*.whl" >/dev/null 2>/dev/null; then 2025-09-09T16:48:24.5473062Z  mv -v dist/*.whl "${RUNNER_ARTIFACT_DIR}/" 2025-09-09T16:48:24.5473362Z  fi 2025-09-09T16:48:24.5473608Z  if [[ -d "artifacts-to-be-uploaded" ]]; then 2025-09-09T16:48:24.5474016Z  mv -v artifacts-to-be-uploaded/* "${RUNNER_ARTIFACT_DIR}/" 2025-09-09T16:48:24.5474368Z  fi 2025-09-09T16:48:24.5488032Z fi 2025-09-09T16:48:24.5488258Z  2025-09-09T16:48:24.5488478Z upload_docs=0 2025-09-09T16:48:24.5488843Z # Check if there are files in the documentation folder to upload, note that 2025-09-09T16:48:24.5489259Z # empty folders do not count 2025-09-09T16:48:24.5489680Z if find "${RUNNER_DOCS_DIR}" -mindepth 1 -maxdepth 1 -type f | read -r; then 2025-09-09T16:48:24.5490363Z  # TODO: Add a check here to test if on ec2 because if we're not on ec2 then this 2025-09-09T16:48:24.5490809Z  # upload will probably not work correctly 2025-09-09T16:48:24.5491164Z  upload_docs=1 2025-09-09T16:48:24.5491384Z fi 2025-09-09T16:48:24.5491654Z echo "upload-docs=${upload_docs}" >> "${GITHUB_OUTPUT}" 2025-09-09T16:48:24.5500643Z shell: /usr/bin/bash -e {0} 2025-09-09T16:48:24.5500956Z env: 2025-09-09T16:48:24.5501195Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:24.5501527Z REPOSITORY: pytorch/ao 2025-09-09T16:48:24.5501761Z PR_NUMBER: 2963 2025-09-09T16:48:24.5503065Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:24.5504518Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:24.5505158Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:24.5505649Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:24.5506061Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:24.5506387Z UPLOAD_ARTIFACT_NAME: 2025-09-09T16:48:24.5506620Z ##[endgroup] 2025-09-09T16:48:24.5609244Z Prepare all required actions 2025-09-09T16:48:24.5643425Z ##[group]Run ./test-infra/.github/actions/teardown-linux 2025-09-09T16:48:24.5643771Z with: 2025-09-09T16:48:24.5643956Z env: 2025-09-09T16:48:24.5644214Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:24.5644559Z REPOSITORY: pytorch/ao 2025-09-09T16:48:24.5644798Z PR_NUMBER: 2963 2025-09-09T16:48:24.5646120Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:24.5647603Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:24.5648136Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:24.5648672Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:24.5649093Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:24.5649440Z ##[endgroup] 2025-09-09T16:48:24.5672364Z ##[group]Run set -eou pipefail 2025-09-09T16:48:24.5672663Z set -eou pipefail 2025-09-09T16:48:24.5672902Z  2025-09-09T16:48:24.5673227Z echo "Holding runner for 2 hours until all ssh sessions have logged out" 2025-09-09T16:48:24.5673644Z for _ in $(seq 1440); do 2025-09-09T16:48:24.5673935Z  # Break if no ssh session exists anymore 2025-09-09T16:48:24.5674249Z  if [ "$(who)" = "" ]; then 2025-09-09T16:48:24.5674503Z  break 2025-09-09T16:48:24.5674712Z  fi 2025-09-09T16:48:24.5674920Z  echo "." 2025-09-09T16:48:24.5675131Z  sleep 5 2025-09-09T16:48:24.5675341Z done 2025-09-09T16:48:24.5684475Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T16:48:24.5684814Z env: 2025-09-09T16:48:24.5685050Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:24.5685375Z REPOSITORY: pytorch/ao 2025-09-09T16:48:24.5685604Z PR_NUMBER: 2963 2025-09-09T16:48:24.5686892Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:24.5688463Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:24.5688991Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:24.5689490Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:24.5689896Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:24.5690221Z ##[endgroup] 2025-09-09T16:48:24.5727555Z Holding runner for 2 hours until all ssh sessions have logged out 2025-09-09T16:48:24.5830200Z ##[group]Run # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T16:48:24.5830744Z # ignore expansion of "docker ps -q" since it could be empty 2025-09-09T16:48:24.5831138Z # shellcheck disable=SC2046 2025-09-09T16:48:24.5831442Z docker stop $(docker ps -q) || true 2025-09-09T16:48:24.5831763Z # Prune all of the docker images 2025-09-09T16:48:24.5832057Z docker system prune -af 2025-09-09T16:48:24.5841336Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T16:48:24.5841683Z env: 2025-09-09T16:48:24.5841944Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:24.5842273Z REPOSITORY: pytorch/ao 2025-09-09T16:48:24.5842534Z PR_NUMBER: 2963 2025-09-09T16:48:24.5844054Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:24.5845519Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:24.5846068Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:24.5846574Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:24.5846986Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:24.5847315Z ##[endgroup] 2025-09-09T16:48:26.5168827Z 6f14d5419571 2025-09-09T16:48:35.4503244Z Deleted Containers: 2025-09-09T16:48:35.4503932Z 6f14d54195719389a6ca04d74d944db4d0b5bcd49ba6fe5591c3d08f98564f8f 2025-09-09T16:48:35.4504424Z 2025-09-09T16:48:43.2831589Z Deleted Images: 2025-09-09T16:48:43.2831993Z untagged: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:43.2832597Z untagged: pytorch/almalinux-builder@sha256:be7f2a4c6f467933b154ac0b3ded894ad1bf06ce95f8f8d908dba108e68806f3 2025-09-09T16:48:43.2833352Z deleted: sha256:f2a1aa8f0f7d42816c0fe19d8838a50734adb5eb5de36e7f04cffdbf8de9d63c 2025-09-09T16:48:43.2833944Z deleted: sha256:c9e150f3add3ad75da73e417a090d5b5fc1b81cc89f87daa049075292636557c 2025-09-09T16:48:43.2834529Z deleted: sha256:a34ecd86569aa00449c46f68a40f7e90b90cc6a7e7d286262816aee915322a7b 2025-09-09T16:48:43.2835110Z deleted: sha256:892ada8183495e70a17e9cfa280133ba3bc836b5e37564e0bbe9d7f8234a3848 2025-09-09T16:48:43.2835701Z deleted: sha256:eadbce3bbb3f1a27ce48c9c43e762dc84723582ebdf7fcaf004049bd5efebcd4 2025-09-09T16:48:43.2836341Z deleted: sha256:614eb17bcea9533c6fafc9f12884904851ce23b0a4609d3e6eb24ebccee3a49d 2025-09-09T16:48:43.2847154Z deleted: sha256:5d7a767e6e9b52e3bd6ae5896a9c32ca564f0880f3bec9eb38678eec35ebf104 2025-09-09T16:48:43.2847756Z deleted: sha256:c375116670abbf5fd150e1ee6ee8f0ea2d14351ec3e6a55e2a897f23ada5cfd6 2025-09-09T16:48:43.2848337Z deleted: sha256:9bd5a8ab14bca745f798e5e88495f278c13d3fc531c4b668c5c8f7c91d334f9b 2025-09-09T16:48:43.2848911Z deleted: sha256:4dc4b770bc6212696f6b304b6c20f9c2c34f899bc1a36629fd3595ea776b846f 2025-09-09T16:48:43.2849480Z deleted: sha256:baf10146ebddd70194ce3977095cc8ed0e6d7df2f88979af5b49262c60a3f3e1 2025-09-09T16:48:43.2850060Z deleted: sha256:4ab11f18df8904dad268820aa8d575115da52680d411d7559440c0b5bc04b0f1 2025-09-09T16:48:43.2850853Z deleted: sha256:3597cd684ef75bc762bbad866d470bde44719232d96ccf3cfcda286583915b14 2025-09-09T16:48:43.2851437Z deleted: sha256:1beac1c8161bfd0d6ac4d67b32f665ccfb1b312f7af82b05d30ac79551275838 2025-09-09T16:48:43.2852030Z deleted: sha256:3391cdf48fdf7b9acbb1ac360ba1210ca8a8d727313aa25f38a977087f547d99 2025-09-09T16:48:43.2852618Z deleted: sha256:45cdedd84abf879e7a255cb152d814c309dbdade3f6a917f6bd54d76277d71d6 2025-09-09T16:48:43.2853209Z deleted: sha256:3e51a87c61a2c818401711cbc3a136cfaa58de4c8453602a012cbdda61e20cd8 2025-09-09T16:48:43.2853778Z deleted: sha256:c494b964f63b3ad12c0b2478a96315d32ddb3d60869a8ff249c5f70b78918f47 2025-09-09T16:48:43.2854359Z deleted: sha256:7dd5e41bb10ca14c9e763dd5cbb9e15c7a4727c5cbc49b40cfd262872ea7b3dd 2025-09-09T16:48:43.2854944Z deleted: sha256:0ebf188609b1c3acf20d30c01cd9d08f623b552df32b6bbd9c9ea72c713cbda7 2025-09-09T16:48:43.2855515Z deleted: sha256:46a0108e49ebe8b5c8a25a046961a83a96de1d1d2b3c83fdb1cce9c24033d3f3 2025-09-09T16:48:43.2856097Z deleted: sha256:add74a43e7c81f71d6040c8c766aa51293da349336c272faccf238aac3218502 2025-09-09T16:48:43.2856813Z deleted: sha256:ebe35679e50c08c29cc5c163b411dcd19c66cbb41714d123f4eaebcd8c836766 2025-09-09T16:48:43.2857658Z deleted: sha256:ea669f8859ecce20d165c51dd89be8761f0ad6f7e899116e386a4f9dd403b565 2025-09-09T16:48:43.2858335Z deleted: sha256:ff4f19608a1944c0c2807cd533515673285a9632dc74bf020e83e18630d1ae35 2025-09-09T16:48:43.2858808Z untagged: public.ecr.aws/docker/library/python:3.13 2025-09-09T16:48:43.2861367Z untagged: public.ecr.aws/docker/library/python@sha256:74503e0bff6cf811f029590a05e0218cc9ba3e099a4b7df0ab84a67df081e1bc 2025-09-09T16:48:43.2862112Z deleted: sha256:77f2b24be2b3987f6d59918787d226acb4e6612644bacb3dd37adc494e477d9e 2025-09-09T16:48:43.2862679Z deleted: sha256:1b9aa91044866f8707424c8fe367f924a48557eac69f7485fd6d2a3a116c74d5 2025-09-09T16:48:43.2863236Z deleted: sha256:b86402d18e73d4825a3bd2a09244a93487ba4687ca7c9dcba0f73e160840845c 2025-09-09T16:48:43.2864139Z deleted: sha256:5755f8963eb047a0086073c3a7dd0731296d6751a7445f3693a52b30020a5b65 2025-09-09T16:48:43.2864698Z deleted: sha256:7f33dbfa9475d25622f49ed51f4164c97de1303331c77dfdc738e084d100f50c 2025-09-09T16:48:43.2865265Z deleted: sha256:19daa38049795ba2c166dd898c81b17e31f4b5f98c1337846c6515fff97d8782 2025-09-09T16:48:43.2865842Z deleted: sha256:483bd23b5d7e66fc0f8a92dbfacc3d72fad97ef47dd4767889979a803bc1f5b8 2025-09-09T16:48:43.2866411Z deleted: sha256:185e04da9d947141fd703dbf36361bdc2ff77cc27cbf500fb9f4881cb5ddbe95 2025-09-09T16:48:43.2866973Z untagged: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine:latest 2025-09-09T16:48:43.2867730Z untagged: 308535385114.dkr.ecr.us-east-1.amazonaws.com/tool/alpine@sha256:def822f9851ca422481ec6fee59a9966f12b351c62ccb9aca841526ffaa9f748 2025-09-09T16:48:43.2868518Z deleted: sha256:6dbb9cc54074106d46d4ccb330f2a40a682d49dda5f4844962b7dce9fe44aaec 2025-09-09T16:48:43.2869104Z deleted: sha256:b2d5eeeaba3a22b9b8aa97261957974a6bd65274ebd43e1d81d0a7b8b752b116 2025-09-09T16:48:43.2869448Z 2025-09-09T16:48:43.2869559Z Total reclaimed space: 30.03GB 2025-09-09T16:48:43.2925719Z ##[group]Run set +e 2025-09-09T16:48:43.2925982Z set +e 2025-09-09T16:48:43.2926222Z if [[ "${NO_SUDO}" == "false" ]]; then 2025-09-09T16:48:43.2926603Z  sudo rm -rf "${GITHUB_WORKSPACE:?}/${REPOSITORY:?}" 2025-09-09T16:48:43.2926931Z else 2025-09-09T16:48:43.2927195Z  rm -rf "${GITHUB_WORKSPACE:?}/${REPOSITORY:?}" 2025-09-09T16:48:43.2927508Z fi 2025-09-09T16:48:43.2927699Z set -e 2025-09-09T16:48:43.2938526Z shell: /usr/bin/bash -e {0} 2025-09-09T16:48:43.2938769Z env: 2025-09-09T16:48:43.2939014Z DOCKER_IMAGE: pytorch/almalinux-builder:cuda12.6 2025-09-09T16:48:43.2939383Z REPOSITORY: pytorch/ao 2025-09-09T16:48:43.2939626Z PR_NUMBER: 2963 2025-09-09T16:48:43.2941130Z SCRIPT: conda create -n venv python=3.9 -y conda activate venv python -m pip install --upgrade pip pip install --pre torch --index-url https://download.pytorch.org/whl/nightly/cu126 pip install -r dev-requirements.txt pip install . export CONDA=$(dirname $(dirname $(which conda))) export LD_LIBRARY_PATH=$CONDA/lib/:$LD_LIBRARY_PATH pytest test --verbose -s 2025-09-09T16:48:43.2942721Z RUNNER_ARTIFACT_DIR: /home/ec2-user/actions-runner/_work/_temp/artifacts 2025-09-09T16:48:43.2943255Z RUNNER_TEST_RESULTS_DIR: /home/ec2-user/actions-runner/_work/_temp/test-results 2025-09-09T16:48:43.2943756Z RUNNER_DOCS_DIR: /home/ec2-user/actions-runner/_work/_temp/docs 2025-09-09T16:48:43.2944174Z GPU_FLAG: --gpus all -e NVIDIA_DRIVER_CAPABILITIES=all 2025-09-09T16:48:43.2944500Z NO_SUDO: false 2025-09-09T16:48:43.2944706Z ##[endgroup] 2025-09-09T16:48:43.7944607Z Post job cleanup. 2025-09-09T16:48:43.9028049Z Post job cleanup. 2025-09-09T16:48:44.0010977Z [command]/usr/bin/git version 2025-09-09T16:48:44.0064010Z git version 2.47.1 2025-09-09T16:48:44.0108138Z Temporarily overriding HOME='/home/ec2-user/actions-runner/_work/_temp/2749bf3f-7494-4d44-9e79-79b896b0f7da' before making global git config changes 2025-09-09T16:48:44.0109016Z Adding repository directory to the temporary git global config as a safe directory 2025-09-09T16:48:44.0113761Z [command]/usr/bin/git config --global --add safe.directory /home/ec2-user/actions-runner/_work/ao/ao/test-infra 2025-09-09T16:48:44.0156813Z [command]/usr/bin/git config --local --name-only --get-regexp core\.sshCommand 2025-09-09T16:48:44.0197427Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'core\.sshCommand' && git config --local --unset-all 'core.sshCommand' || :" 2025-09-09T16:48:44.0615766Z [command]/usr/bin/git config --local --name-only --get-regexp http\.https\:\/\/github\.com\/\.extraheader 2025-09-09T16:48:44.0645084Z http.https://github.com/.extraheader 2025-09-09T16:48:44.0657241Z [command]/usr/bin/git config --local --unset-all http.https://github.com/.extraheader 2025-09-09T16:48:44.0694883Z [command]/usr/bin/git submodule foreach --recursive sh -c "git config --local --name-only --get-regexp 'http\.https\:\/\/github\.com\/\.extraheader' && git config --local --unset-all 'http.https://github.com/.extraheader' || :" 2025-09-09T16:48:44.1186631Z A job completed hook has been configured by the self-hosted runner administrator 2025-09-09T16:48:44.1219415Z ##[group]Run '/home/ec2-user/runner-scripts/after_job.sh' 2025-09-09T16:48:44.1227446Z shell: /usr/bin/bash --noprofile --norc -e -o pipefail {0} 2025-09-09T16:48:44.1227796Z ##[endgroup] 2025-09-09T16:48:44.1347224Z [!ALERT!] Swap in detected! [!ALERT!] 2025-09-09T16:48:55.4625519Z [!ALERT!] Swap out detected [!ALERT!] 2025-09-09T16:49:13.6774954Z Cleaning up orphan processes