Skip to content

Upgrade Big Model pipeline CUDA from 11.8 to 12.x #24156

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
tianleiwu merged 2 commits into from
Mar 25, 2025
Merged

Upgrade Big Model pipeline CUDA from 11.8 to 12.x #24156

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
8d1f7be
use cuda 12
tianleiwu Mar 25, 2025
36236f4
torch for cu124
tianleiwu Mar 25, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
27 changes: 14 additions & 13 deletions tools/ci_build/github/azure-pipelines/bigmodels-ci-pipeline.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,11 +41,11 @@ parameters:

variables:
- name: docker_base_image
value: onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntime/build/cuda11_x64_almalinux8_gcc11:20250124.1
value: onnxruntimebuildcache.azurecr.io/internal/azureml/onnxruntime/build/cuda12_x64_ubi8_gcc12:20250124.1
- name: linux_trt_version
value: 10.3.0.26-1.cuda11.8
value: 10.9.0.34-1.cuda12.8
- name: Repository
value: 'onnxruntimecuda11manylinuxbuild'
value: 'onnxruntimecuda12manylinuxbuild'

stages:
- stage: Build_Onnxruntime_Cuda
Expand Down Expand Up @@ -93,16 +93,17 @@ stages:
$(Repository) \
/bin/bash -c '
set -ex; \
ls /usr/local/; \
PATH=/opt/python/cp310-cp310/bin:$PATH /opt/python/cp310-cp310/bin/python3 /onnxruntime_src/tools/ci_build/build.py \
--build_dir /build --cmake_generator Ninja \
--config Release --update --build \
--skip_submodule_sync \
--build_shared_lib \
--parallel --use_vcpkg --use_vcpkg_ms_internal_asset_cache \
--build_wheel \
--enable_onnx_tests --use_cuda --cuda_version=11.8 --cuda_home=/usr/local/cuda-11.8 --cudnn_home=/usr/local/cuda-11.8 \
--enable_onnx_tests --use_cuda --cuda_version=12.2 --cuda_home=/usr/local/cuda-12.2 --cudnn_home=/usr/local/cuda-12.2 \
--enable_cuda_profiling \
--enable_pybind --build_java \
--enable_pybind \
--cmake_extra_defines "CMAKE_CUDA_ARCHITECTURES=75;86" '

workingDirectory: $(Build.SourcesDirectory)
Expand Down Expand Up @@ -164,7 +165,7 @@ stages:
DockerBuildArgs: "
--build-arg BUILD_UID=$( id -u )
"
Repository: onnxruntimeubuntupackagestest_cuda11
Repository: onnxruntimeubuntupackagestest_cuda12
UseImageCacheContainerRegistry: false

- task: Cache@2
Expand All @@ -182,14 +183,14 @@ stages:
-v $(Build.BinariesDirectory)/Release:/Release \
-v $(STABLE_DIFFUSION_MODEL_CACHE):/model_cache:rw \
-v $(GenerateImage_DIR):/images:rw \
onnxruntimeubuntupackagestest_cuda11 \
onnxruntimeubuntupackagestest_cuda12 \
bash -c ' \
set -ex; \
python3 --version; \
python3 -m pip install --upgrade pip; \
python3 -m pip install /Release/*.whl; \
pushd /workspace/onnxruntime/python/tools/transformers/models/stable_diffusion; \
python3 -m pip install -r requirements/cuda11/requirements.txt; \
python3 -m pip install -r requirements/cuda12/requirements.txt; \
python3 -m pip install numpy==1.22.2; \
python3 -m pip install --upgrade polygraphy onnx-graphsurgeon ; \
echo Generate an image guided by a text prompt; \
Expand Down Expand Up @@ -221,7 +222,7 @@ stages:
- script: |
docker run -e SYSTEM_COLLECTIONURI --rm --gpus all -v $PWD:/workspace \
-v $(CLIP_MODEL_CACHE):/model_cache:rw \
onnxruntimeubuntupackagestest_cuda11 \
onnxruntimeubuntupackagestest_cuda12 \
bash -c '
set -x; \
python3 --version; \
Expand All @@ -248,7 +249,7 @@ stages:
- script: |
docker run -e SYSTEM_COLLECTIONURI --rm --gpus all -v $PWD:/workspace \
-v $(CLIP_MODEL_CACHE):/model_cache:rw \
onnxruntimeubuntupackagestest_cuda11 \
onnxruntimeubuntupackagestest_cuda12 \
bash -c '
set -ex; \
python3 --version; \
Expand Down Expand Up @@ -445,7 +446,7 @@ stages:
popd ; \
python3 -m pip install /ort-artifact/*.whl ; \
python3 -m pip uninstall -y torch ; \
python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu118 ; \
python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu124 ; \
python3 -m models.whisper.convert_to_onnx -m openai/whisper-tiny --output wtiny-fp32-cpu-hf --precision fp32 --provider cpu --overwrite --use_external_data_format --optimize_onnx --no_beam_search_op --output_cross_qk ; \
python3 -m models.whisper.convert_to_onnx -m openai/whisper-tiny --output wtiny-fp32-cpu-hf --precision fp32 --provider cpu --overwrite --use_external_data_format --optimize_onnx ; \
python3 -m models.whisper.convert_to_onnx -m openai/whisper-tiny --output wtiny-fp32-cpu-hf --precision fp32 --provider cpu --overwrite --use_external_data_format --optimize_onnx --no_beam_search_op --output_cross_qk --separate_encoder_and_decoder_init ; \
Expand Down Expand Up @@ -489,7 +490,7 @@ stages:
popd ; \
python3 -m pip install /ort-artifact/*.whl ; \
python3 -m pip uninstall -y torch ; \
python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu118 ; \
python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu124 ; \
python3 -m models.whisper.convert_to_onnx -m /whisper_large_v3 --output whisperlargev3 --use_external_data_format ; \
popd ; \
'
Expand All @@ -510,7 +511,7 @@ stages:
popd ; \
python3 -m pip install /ort-artifact/*.whl ; \
python3 -m pip uninstall -y torch ; \
python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu118 ; \
python3 -m pip install torch --index-url https://download.pytorch.org/whl/cu124 ; \
ls whisperlargev3; \
export LD_LIBRARY_PATH=/tmp/ompffmpeg:${LD_LIBRARY_PATH}; \
ffmpeg -version; \
Expand Down
6 changes: 3 additions & 3 deletions tools/ci_build/github/linux/docker/Dockerfile.package_ubuntu_2204_gpu_ffmpeg
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,8 @@
# Dockerfile to run ONNXRuntime with TensorRT integration

# Build base image with required system packages
ARG BASEIMAGE=nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
ARG TRT_VERSION=10.9.0.34-1+cuda11.8
ARG BASEIMAGE=nvidia/cuda:12.8.1-cudnn-devel-ubuntu22.04
ARG TRT_VERSION=10.9.0.34-1+cuda12.8
ARG LD_LIBRARY_PATH_ARG=/usr/local/lib64:/usr/local/cuda/lib64
FROM $BASEIMAGE AS base
ARG TRT_VERSION
Expand All @@ -28,7 +28,7 @@ RUN apt-get install -y --no-install-recommends \
RUN pip install --upgrade pip

# Install TensorRT
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub &&\
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/7fa2af80.pub &&\
apt-get update &&\
apt-get install -y \
libnvinfer-dev=${TRT_VERSION} \
Expand Down
6 changes: 3 additions & 3 deletions tools/ci_build/github/linux/docker/Dockerfile.package_ubuntu_2204_gpu_opencv
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -5,8 +5,8 @@
# Dockerfile to run ONNXRuntime with TensorRT integration

# Build base image with required system packages
ARG BASEIMAGE=nvidia/cuda:11.8.0-cudnn8-devel-ubuntu22.04
ARG TRT_VERSION=10.9.0.34-1+cuda11.8
ARG BASEIMAGE=nvidia/cuda:12.8.1-cudnn-devel-ubuntu22.04
ARG TRT_VERSION=10.9.0.34-1+cuda12.8
ARG LD_LIBRARY_PATH_ARG=/usr/local/lib64:/usr/local/cuda/lib64
FROM $BASEIMAGE AS base
ARG TRT_VERSION
Expand All @@ -30,7 +30,7 @@ RUN apt-get install -y --no-install-recommends \
RUN pip install --upgrade pip

# Install TensorRT
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/x86_64/7fa2af80.pub &&\
RUN apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2204/x86_64/7fa2af80.pub &&\
apt-get update &&\
apt-get install -y \
libnvinfer-dev=${TRT_VERSION} \
Expand Down
Loading