Skip to content

[DRAFT] Optimize Dockerfile to reduce image size and build time. #1226

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

Open
py4 wants to merge 3 commits into
base: main
Choose a base branch
from
Open

[DRAFT] Optimize Dockerfile to reduce image size and build time. #1226

Changes from all commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
d746104
Optimize Dockerfile to reduce image size and build time.
Dec 3, 2025
cba1768
Merge branch 'main' into pooyam4
py4 Dec 3, 2025
5b1d688
Merge branch 'main' into pooyam4
py4 Dec 3, 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
31 changes: 21 additions & 10 deletions docker/Dockerfile
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -11,9 +11,12 @@ ARG IS_FOR_V7X="false"
RUN pip uninstall -y torch torch_xla torchvision

# Install some basic utilities
RUN apt-get update && apt-get install -y \
RUN --mount=type=cache,target=/var/cache/apt,sharing=locked \
--mount=type=cache,target=/var/lib/apt,sharing=locked \
apt-get update && apt-get install -y \
git \
libopenblas-base libopenmpi-dev libomp-dev
libopenblas-base libopenmpi-dev libomp-dev \
&& rm -rf /var/lib/apt/lists/*

# Build vLLM
WORKDIR /workspace/vllm
Expand All @@ -23,12 +26,16 @@ RUN git clone $VLLM_REPO /workspace/vllm
RUN if [ -n "$VLLM_COMMIT_HASH" ]; then \
git checkout $VLLM_COMMIT_HASH; \
fi
RUN pip install -r requirements/tpu.txt --retries 3
RUN VLLM_TARGET_DEVICE="tpu" pip install -e .
RUN --mount=type=cache,target=/root/.cache/pip \
pip install -r requirements/tpu.txt --retries 3
RUN --mount=type=cache,target=/root/.cache/pip \
VLLM_TARGET_DEVICE="tpu" pip install -e .

# Install test dependencies
RUN python3 -m pip install -e tests/vllm_test_utils
RUN python3 -m pip install --no-cache-dir \
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install -e tests/vllm_test_utils
RUN --mount=type=cache,target=/root/.cache/pip \
python3 -m pip install \
git+https://github.com/thuml/depyf.git \
pytest-asyncio \
git+https://github.com/EleutherAI/lm-evaluation-harness.git@206b7722158f58c35b7ffcd53b035fdbdda5126d#egg=lm-eval[api] \
Expand All @@ -39,18 +46,22 @@ RUN python3 -m pip install --no-cache-dir \
WORKDIR /workspace/tpu_inference
# Install requirements first and cache so we don't need to re-install on code change.
COPY requirements.txt .
RUN pip install -r requirements.txt --retries 3
RUN --mount=type=cache,target=/root/.cache/pip \
pip install -r requirements.txt --retries 3
COPY requirements_benchmarking.txt .
# These are needed for the E2E benchmarking tests (i.e. tests/e2e/benchmarking/mlperf.sh)
RUN pip install -r requirements_benchmarking.txt --retries 3
RUN --mount=type=cache,target=/root/.cache/pip \
pip install -r requirements_benchmarking.txt --retries 3
COPY . .
RUN pip install -e .
RUN --mount=type=cache,target=/root/.cache/pip \
pip install -e .

# TODO (jacobplatin): remove when v7x is supported in JAX/Libtpu officially
# NOTE: it's important that this is done after installing tpu_inference above,
# so that the v7x-specific dependencies can override any existing ones.
COPY requirements_v7x.txt .
RUN if [ "$IS_FOR_V7X" = "true" ]; then \
RUN --mount=type=cache,target=/root/.cache/pip \
if [ "$IS_FOR_V7X" = "true" ]; then \
pip install -r requirements_v7x.txt; \
fi

Expand Down