Skip to content

[CI/Build] Use uv in the Dockerfile #13566

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
merged 2 commits into from
Feb 20, 2025
Merged

[CI/Build] Use uv in the Dockerfile #13566

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
73102bb
[CI/Build] Use uv in the Dockerfile
mgoin Feb 19, 2025
cf28e17
Updates
mgoin Feb 19, 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
32 changes: 19 additions & 13 deletions Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,6 +27,9 @@ RUN echo 'tzdata tzdata/Areas select America' | debconf-set-selections \
&& ln -sf /usr/bin/python${PYTHON_VERSION}-config /usr/bin/python3-config \
&& curl -sS https://bootstrap.pypa.io/get-pip.py | python${PYTHON_VERSION} \
&& python3 --version && python3 -m pip --version
# Install uv for faster pip installs
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install uv

# Upgrade to GCC 10 to avoid https://gcc.gnu.org/bugzilla/show_bug.cgi?id=92519
# as it was causing spam when compiling the CUTLASS kernels
Expand All @@ -52,13 +55,13 @@ WORKDIR /workspace
# after this step
RUN --mount=type=cache,target=/root/.cache/pip \
if [ "$TARGETPLATFORM" = "linux/arm64" ]; then \
python3 -m pip install --index-url https://download.pytorch.org/whl/nightly/cu126 "torch==2.7.0.dev20250121+cu126" "torchvision==0.22.0.dev20250121"; \
uv pip install --system --index-url https://download.pytorch.org/whl/nightly/cu126 "torch==2.7.0.dev20250121+cu126" "torchvision==0.22.0.dev20250121"; \
fi

COPY requirements-common.txt requirements-common.txt
COPY requirements-cuda.txt requirements-cuda.txt
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -r requirements-cuda.txt
uv pip install --system -r requirements-cuda.txt

# cuda arch list used by torch
# can be useful for both `dev` and `test`
Expand All @@ -79,7 +82,7 @@ ARG TARGETPLATFORM
COPY requirements-build.txt requirements-build.txt

RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -r requirements-build.txt
uv pip install --system -r requirements-build.txt

COPY . .
ARG GIT_REPO_CHECK=0
Expand Down Expand Up @@ -144,7 +147,7 @@ COPY requirements-lint.txt requirements-lint.txt
COPY requirements-test.txt requirements-test.txt
COPY requirements-dev.txt requirements-dev.txt
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -r requirements-dev.txt
uv pip install --system -r requirements-dev.txt
#################### DEV IMAGE ####################

#################### vLLM installation IMAGE ####################
Expand Down Expand Up @@ -174,6 +177,9 @@ RUN echo 'tzdata tzdata/Areas select America' | debconf-set-selections \
&& ln -sf /usr/bin/python${PYTHON_VERSION}-config /usr/bin/python3-config \
&& curl -sS https://bootstrap.pypa.io/get-pip.py | python${PYTHON_VERSION} \
&& python3 --version && python3 -m pip --version
# Install uv for faster pip installs
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install uv

# Workaround for https://github.com/openai/triton/issues/2507 and
# https://github.com/pytorch/pytorch/issues/107960 -- hopefully
Expand All @@ -187,13 +193,13 @@ RUN ldconfig /usr/local/cuda-$(echo $CUDA_VERSION | cut -d. -f1,2)/compat/
# after this step
RUN --mount=type=cache,target=/root/.cache/pip \
if [ "$TARGETPLATFORM" = "linux/arm64" ]; then \
python3 -m pip install --index-url https://download.pytorch.org/whl/nightly/cu124 "torch==2.6.0.dev20241210+cu124" "torchvision==0.22.0.dev20241215"; \
uv pip install --system --index-url https://download.pytorch.org/whl/nightly/cu124 "torch==2.6.0.dev20241210+cu124" "torchvision==0.22.0.dev20241215"; \
fi

# Install vllm wheel first, so that torch etc will be installed.
RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist \
--mount=type=cache,target=/root/.cache/pip \
python3 -m pip install dist/*.whl --verbose
uv pip install --system dist/*.whl --verbose

# If we need to build FlashInfer wheel before its release:
# $ export FLASHINFER_ENABLE_AOT=1
Expand All @@ -210,7 +216,7 @@ RUN --mount=type=bind,from=build,src=/workspace/dist,target=/vllm-workspace/dist
RUN --mount=type=cache,target=/root/.cache/pip \
. /etc/environment && \
if [ "$TARGETPLATFORM" != "linux/arm64" ]; then \
python3 -m pip install https://github.com/flashinfer-ai/flashinfer/releases/download/v0.2.1.post1/flashinfer_python-0.2.1.post1+cu124torch2.5-cp38-abi3-linux_x86_64.whl ; \
uv pip install --system https://github.com/flashinfer-ai/flashinfer/releases/download/v0.2.1.post1/flashinfer_python-0.2.1.post1+cu124torch2.5-cp38-abi3-linux_x86_64.whl ; \
fi
COPY examples examples

Expand All @@ -220,7 +226,7 @@ COPY examples examples
# TODO: Remove this once FlashInfer AOT wheel is fixed
COPY requirements-build.txt requirements-build.txt
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -r requirements-build.txt
uv pip install --system -r requirements-build.txt

#################### vLLM installation IMAGE ####################

Expand All @@ -233,15 +239,15 @@ ADD . /vllm-workspace/

# install development dependencies (for testing)
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -r requirements-dev.txt
uv pip install --system -r requirements-dev.txt

# install development dependencies (for testing)
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -e tests/vllm_test_utils
uv pip install --system -e tests/vllm_test_utils

# enable fast downloads from hf (for testing)
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install hf_transfer
uv pip install --system hf_transfer
ENV HF_HUB_ENABLE_HF_TRANSFER 1

# Copy in the v1 package for testing (it isn't distributed yet)
Expand All @@ -262,9 +268,9 @@ FROM vllm-base AS vllm-openai-base
# install additional dependencies for openai api server
RUN --mount=type=cache,target=/root/.cache/pip \
if [ "$TARGETPLATFORM" = "linux/arm64" ]; then \
pip install accelerate hf_transfer 'modelscope!=1.15.0' 'bitsandbytes>=0.42.0' 'timm==0.9.10' boto3 runai-model-streamer runai-model-streamer[s3]; \
uv pip install --system accelerate hf_transfer 'modelscope!=1.15.0' 'bitsandbytes>=0.42.0' 'timm==0.9.10' boto3 runai-model-streamer runai-model-streamer[s3]; \
else \
pip install accelerate hf_transfer 'modelscope!=1.15.0' 'bitsandbytes>=0.45.0' 'timm==0.9.10' boto3 runai-model-streamer runai-model-streamer[s3]; \
uv pip install --system accelerate hf_transfer 'modelscope!=1.15.0' 'bitsandbytes>=0.45.0' 'timm==0.9.10' boto3 runai-model-streamer runai-model-streamer[s3]; \
fi

ENV VLLM_USAGE_SOURCE production-docker-image
Expand Down