sha256:904520f3fefc32414abcedf6ba8dadc9c5978db1bcdaf40f58d6b9036ef30a99
OS/ARCH
Compressed size
9.41 GB
Last pushed
16 days by vllmbot
Type
Image
Manifest digest
sha256:88175ddf4f59944ccbb24c6b60c1bc535a160be3c4f09a3b08702c5e301a599e
Image Layers
1ARG RELEASE0 B2ARG LAUNCHPAD_BUILD_ARCH0 B3LABEL org.opencontainers.image.version=24.040 B4ADD file ... in / 29.18 MB5CMD ["/bin/bash"]0 B6ENV NVARCH=x86_640 B7ENV NVIDIA_REQUIRE_CUDA=cuda>=13.0 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<5360 B8ENV NV_CUDA_CUDART_VERSION=13.0.96-10 B9ARG TARGETARCH0 B11RUN |1 TARGETARCH=amd64 /bin/sh -c6.64 MB12ENV CUDA_VERSION=13.0.30 B13RUN |1 TARGETARCH=amd64 /bin/sh -c354.02 KB14RUN |1 TARGETARCH=amd64 /bin/sh -c100.03 MB15RUN |1 TARGETARCH=amd64 /bin/sh -c32 B16RUN |1 TARGETARCH=amd64 /bin/sh -c186 B17ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B18ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B19COPY NGC-DL-CONTAINER-LICENSE / # buildkit6.72 KB20ENV NVIDIA_VISIBLE_DEVICES=all0 B21ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility0 B22ARG INSTALL_RUBIN_PRERELEASE=false0 B23ARG TARGETPLATFORM=linux/amd640 B24ARG max_jobs=640 B25ARG CUDA_VERSION=13.0.30 B26ARG PYTHON_VERSION=3.120 B27ARG NCCL_VERSION=2.31.20 B28ARG DEADSNAKES_MIRROR_URL0 B29ARG DEADSNAKES_GPGKEY_URL0 B30ARG GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py0 B31ENV DEBIAN_FRONTEND=noninteractive0 B32WORKDIR /vllm-workspace104 B33RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64229 B34RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64221.83 MB35RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=641.35 GB36ENV CUDA_HOME=/usr/local/cuda0 B37ENV PATH=/usr/local/cuda/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B38ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B39COPY tools/install_snapshot_runtime.sh /usr/local/bin/install_snapshot_runtime.sh # buildkit1.62 KB40RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=648.85 MB41ENV CRIU_CUDA_PLUGIN_DIR=/usr/local/lib/criu0 B42RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6433.71 MB43ENV UV_HTTP_TIMEOUT=5000 B44ENV UV_INDEX_STRATEGY=unsafe-best-match0 B45ENV UV_LINK_MODE=copy0 B46ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python0 B47ENV UV_CACHE_DIR=/opt/uv/cache0 B48ENV UV_COMPILE_BYTECODE=10 B49RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64149 B50RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64164 B51ENV UV_OVERRIDE=/etc/uv-overrides.txt0 B52RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=643.42 KB53COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB54RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6432 B55ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B56ARG PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl0 B57COPY requirements/common.txt /tmp/common.txt # buildkit1.7 KB58COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit988 B59RUN |10 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=643.9 GB60RUN |10 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=643.07 GB61ARG GDRCOPY_CUDA_VERSION=12.80 B62ARG GDRCOPY_OS_VERSION=Ubuntu24_040 B63ARG TARGETPLATFORM=linux/amd640 B64COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit897 B65RUN |12 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64336.61 KB66ARG TIMM_VERSION=>=1.0.170 B67ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B68ARG GGUF_VERSION=>=0.19.00 B69RUN |15 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6463.27 MB70ARG PIP_INDEX_URL UV_INDEX_URL=0 B71ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL=0 B72ARG PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl0 B73ARG PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled0 B74ARG PYTORCH_NIGHTLY0 B75COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit196 B76COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit235 B77RUN |22 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64333.72 MB78RUN |22 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64201 B79RUN |22 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6432 B80ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B81COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit253 B82RUN |22 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6415.67 MB83ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B84COPY examples examples # buildkit316.73 KB85COPY benchmarks benchmarks # buildkit312.98 KB86COPY ./vllm/collect_env.py . # buildkit9.61 KB87ARG TARGETPLATFORM=linux/amd640 B88ARG INSTALL_KV_CONNECTORS=true0 B89ARG INSTALL_RUBIN_PRERELEASE=false0 B90ARG TRITON_INSTALL_FROM_SOURCE_REPO=0 B91ARG CUDA_VERSION=13.0.30 B92ARG VLLM_BUILD_COMMIT0 B93ARG VLLM_BUILD_PIPELINE0 B94ARG VLLM_BUILD_URL0 B95ARG VLLM_IMAGE_TAG0 B96ARG PIP_INDEX_URL UV_INDEX_URL=0 B97ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL=0 B98ENV UV_HTTP_TIMEOUT=5000 B99ARG torch_cuda_arch_list=8.7 8.9 9.0 10.0+PTX0 B100ENV TORCH_CUDA_ARCH_LIST=8.7 8.9 9.0 10.0+PTX0 B101RUN |29 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64295.51 MB102RUN |29 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6432 B103COPY /tmp/triton_from_source_workspace/dist/wheel.sha256 /tmp/triton-source-wheel.sha256 # buildkit212 B104COPY tools/install_triton_from_source.sh /tmp/install_triton_from_source.sh # buildkit824 B105RUN |29 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=64174 B106ARG BUILD_NIXL=false0 B107RUN |30 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=6432 B108ENV VLLM_USAGE_SOURCE=production-docker-image0 B109ENV VLLM_BUILD_COMMIT=unknown VLLM_BUILD_PIPELINE=local VLLM_BUILD_URL= VLLM_IMAGE_TAG=local/vllm-openai:dev0 B110LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=unknown org.opencontainers.image.version=local/vllm-openai:dev org.opencontainers.image.url=0 B111ENTRYPOINT ["vllm" "serve"]0 BCommand
ARG RELEASE