co42/crtest-bench-vllm:zstd

Manifest digest

d34b85222203

Image size

7.3 GB

Last pushed

about 1 month ago

Type

application/vnd.oci.image.manifest.v1+json

Image Layers

  1. 1 ARG RELEASE 0 B
  2. 2 ARG LAUNCHPAD_BUILD_ARCH 0 B
  3. 3 LABEL org.opencontainers.image.ref.name=ubuntu 0 B
  4. 4 LABEL org.opencontainers.image.version=24.04 0 B
  5. 5 ADD file:249778a1782b02a1c2bcf9f292f5778d81442a53c3de1958d712f10baf7e0b60 in / 27.3 MB
  6. 6 CMD ["/bin/bash"] 0 B
  7. 7 ENV NVARCH=x86_64 0 B
  8. 8 ENV NVIDIA_REQUIRE_CUDA=cuda>=13.0 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<536 brand=nvidia,driver>=535,driver<536 brand=quadro,driver>=535,driver<536 brand=quadrortx,driver>=535,driver<536 brand=nvidiartx,driver>=535,driver<536 brand=vapps,driver>=535,driver<536 brand=vpc,driver>=535,driver<536 brand=vcs,driver>=535,driver<536 brand=vws,driver>=535,driver<536 brand=cloudgaming,driver>=535,driver<536 brand=unknown,driver>=550,driver<551 brand=grid,driver>=550,driver<551 brand=tesla,driver>=550,driver<551 brand=nvidia,driver>=550,driver<551 brand=quadro,driver>=550,driver<551 brand=quadrortx,driver>=550,driver<551 brand=nvidiartx,driver>=550,driver<551 brand=vapps,driver>=550,driver<551 brand=vpc,driver>=550,driver<551 brand=vcs,driver>=550,driver<551 brand=vws,driver>=550,driver<551 brand=cloudgaming,driver>=550,driver<551 brand=unknown,driver>=565,driver<566 brand=grid,driver>=565,driver<566 brand=tesla,driver>=565,driver<566 brand=nvidia,driver>=565,driver<566 brand=quadro,driver>=565,driver<566 brand=quadrortx,driver>=565,driver<566 brand=nvidiartx,driver>=565,driver<566 brand=vapps,driver>=565,driver<566 brand=vpc,driver>=565,driver<566 brand=vcs,driver>=565,driver<566 brand=vws,driver>=565,driver<566 brand=cloudgaming,driver>=565,driver<566 brand=unknown,driver>=570,driver<571 brand=grid,driver>=570,driver<571 brand=tesla,driver>=570,driver<571 brand=nvidia,driver>=570,driver<571 brand=quadro,driver>=570,driver<571 brand=quadrortx,driver>=570,driver<571 brand=nvidiartx,driver>=570,driver<571 brand=vapps,driver>=570,driver<571 brand=vpc,driver>=570,driver<571 brand=vcs,driver>=570,driver<571 brand=vws,driver>=570,driver<571 brand=cloudgaming,driver>=570,driver<571 brand=unknown,driver>=575,driver<576 brand=grid,driver>=575,driver<576 brand=tesla,driver>=575,driver<576 brand=nvidia,driver>=575,driver<576 brand=quadro,driver>=575,driver<576 brand=quadrortx,driver>=575,driver<576 brand=nvidiartx,driver>=575,driver<576 brand=vapps,driver>=575,driver<576 brand=vpc,driver>=575,driver<576 brand=vcs,driver>=575,driver<576 brand=vws,driver>=575,driver<576 brand=cloudgaming,driver>=575,driver<576 0 B
  9. 9 ENV NV_CUDA_CUDART_VERSION=13.0.96-1 0 B
  10. 10 ARG TARGETARCH 0 B
  11. 11 LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com> 0 B
  12. 12 RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends gnupg2 curl ca-certificates && curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/${NVARCH}/3bf863cc.pub | apt-key add - && echo "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2404/${NVARCH} /" > /etc/apt/sources.list.d/cuda.list && apt-get purge --autoremove -y curl && rm -rf /var/lib/apt/lists/* # buildkit 4.32 MB
  13. 13 ENV CUDA_VERSION=13.0.2 0 B
  14. 14 RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends cuda-cudart-13-0=${NV_CUDA_CUDART_VERSION} cuda-compat-13-0 && rm -rf /var/lib/apt/lists/* # buildkit 95.1 MB
  15. 15 RUN |1 TARGETARCH=amd64 /bin/sh -c echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit 174 Bytes
  16. 16 ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin 0 B
  17. 17 ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64 0 B
  18. 18 COPY NGC-DL-CONTAINER-LICENSE / # buildkit 7.21 kB
  19. 19 ENV NVIDIA_VISIBLE_DEVICES=all 0 B
  20. 20 ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility 0 B
  21. 21 ARG INSTALL_RUBIN_PRERELEASE 0 B
  22. 22 ARG TARGETPLATFORM 0 B
  23. 23 ARG max_jobs=2 0 B
  24. 24 ARG CUDA_VERSION 0 B
  25. 25 ARG PYTHON_VERSION 0 B
  26. 26 ARG NCCL_VERSION 0 B
  27. 27 ARG DEADSNAKES_MIRROR_URL 0 B
  28. 28 ARG DEADSNAKES_GPGKEY_URL 0 B
  29. 29 ARG GET_PIP_URL 0 B
  30. 30 ENV DEBIAN_FRONTEND=noninteractive 0 B
  31. 31 WORKDIR /vllm-workspace 100 Bytes
  32. 32 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c PYTHON_VERSION_STR=$(echo ${PYTHON_VERSION} | sed 's/\.//g') && echo "export PYTHON_VERSION_STR=${PYTHON_VERSION_STR}" >> /etc/environment # buildkit 229 Bytes
  33. 33 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c apt-get update -y && apt-get install -y --no-install-recommends software-properties-common curl sudo ffmpeg libsm6 libxext6 libgl1 && if [ ! -z ${DEADSNAKES_MIRROR_URL} ] ; then if [ ! -z "${DEADSNAKES_GPGKEY_URL}" ] ; then mkdir -p -m 0755 /etc/apt/keyrings ; curl -L ${DEADSNAKES_GPGKEY_URL} | gpg --dearmor > /etc/apt/keyrings/deadsnakes.gpg ; sudo chmod 644 /etc/apt/keyrings/deadsnakes.gpg ; echo "deb [signed-by=/etc/apt/keyrings/deadsnakes.gpg] ${DEADSNAKES_MIRROR_URL} $(lsb_release -cs) main" > /etc/apt/sources.list.d/deadsnakes.list ; fi ; else for i in 1 2 3; do add-apt-repository -y ppa:deadsnakes/ppa && break || { echo "Attempt $i failed, retrying in 5s..."; sleep 5; }; done ; fi && apt-get update -y && apt-get install -y --no-install-recommends python${PYTHON_VERSION} python${PYTHON_VERSION}-dev python${PYTHON_VERSION}-venv libibverbs-dev && rm -rf /var/lib/apt/lists/* && update-alternatives --install /usr/bin/python3 python3 /usr/bin/python${PYTHON_VERSION} 1 && update-alternatives --set python3 /usr/bin/python${PYTHON_VERSION} && ln -sf /usr/bin/python${PYTHON_VERSION}-config /usr/bin/python3-config && rm -f /usr/lib/python${PYTHON_VERSION}/EXTERNALLY-MANAGED && curl -sS ${GET_PIP_URL} | python${PYTHON_VERSION} && python3 --version && python3 -m pip --version # buildkit 220 MB
  34. 34 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-') && apt-get update -y && apt-get install -y --no-install-recommends --allow-change-held-packages cuda-nvcc-${CUDA_VERSION_DASH} cuda-cudart-${CUDA_VERSION_DASH} cuda-nvrtc-${CUDA_VERSION_DASH} cuda-nvrtc-dev-${CUDA_VERSION_DASH} cuda-cuobjdump-${CUDA_VERSION_DASH} libcurand-dev-${CUDA_VERSION_DASH} libcublas-dev-${CUDA_VERSION_DASH} libnuma-dev numactl && rm -rf /var/lib/apt/lists/* # buildkit 1.36 GB
  35. 35 ENV CUDA_HOME=/usr/local/cuda 0 B
  36. 36 ENV PATH=/usr/local/cuda/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin 0 B
  37. 37 ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64 0 B
  38. 38 COPY tools/install_snapshot_runtime.sh /usr/local/bin/install_snapshot_runtime.sh # buildkit 1.73 kB
  39. 39 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c if [ "${TARGETPLATFORM}" = "linux/amd64" ] && [ "${CUDA_VERSION%%.*}" -ge 13 ]; then MAX_JOBS="${max_jobs}" /usr/local/bin/install_snapshot_runtime.sh; else echo "Initialized snapshots are not installed for ${TARGETPLATFORM} with CUDA ${CUDA_VERSION}"; fi # buildkit 8.76 MB
  40. 40 ENV CRIU_CUDA_PLUGIN_DIR=/usr/local/lib/criu 0 B
  41. 41 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c python3 -m pip install uv # buildkit 34.5 MB
  42. 42 ENV UV_HTTP_TIMEOUT=500 0 B
  43. 43 ENV UV_INDEX_STRATEGY=unsafe-best-match 0 B
  44. 44 ENV UV_LINK_MODE=copy 0 B
  45. 45 ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python 0 B
  46. 46 ENV UV_CACHE_DIR=/opt/uv/cache 0 B
  47. 47 ENV UV_COMPILE_BYTECODE=1 0 B
  48. 48 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c mkdir -p "${UV_PYTHON_INSTALL_DIR}" "${UV_CACHE_DIR}" && chgrp -R 0 /opt/uv && chmod -R g+rwX,a+rX /opt/uv # buildkit 149 Bytes
  49. 49 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c CUDA_MAJOR=$(echo $CUDA_VERSION | cut -d. -f1) && if [ -n "$NCCL_VERSION" ]; then echo "nvidia-nccl-cu${CUDA_MAJOR}==${NCCL_VERSION}" > /etc/uv-overrides.txt; else touch /etc/uv-overrides.txt; fi # buildkit 170 Bytes
  50. 50 ENV UV_OVERRIDE=/etc/uv-overrides.txt 0 B
  51. 51 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c useradd --uid 2000 --gid 0 --create-home --home-dir /home/vllm --shell /bin/bash vllm && mkdir -p /home/vllm/.cache /home/vllm/.config && chown -R 2000:0 /home/vllm && chmod -R g+rwX /home/vllm && chgrp 0 /etc/passwd /etc/group && chmod g=u /etc/passwd /etc/group # buildkit 3.79 kB
  52. 52 COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit 2.2 kB
  53. 53 RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c chmod 0755 /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit 2.2 kB
  54. 54 ENV VLLM_ENABLE_CUDA_COMPATIBILITY=0 0 B
  55. 55 ARG PYTORCH_CUDA_INDEX_BASE_URL 0 B
  56. 56 COPY requirements/common.txt /tmp/common.txt # buildkit 1.79 kB
  57. 57 COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit 742 Bytes
  58. 58 RUN |10 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl /bin/sh -c if [ "${INSTALL_RUBIN_PRERELEASE}" = "true" ] && [ ! -f /tmp/vllm-requirements/rubin-prerelease.txt ]; then echo "INSTALL_RUBIN_PRERELEASE=true requires requirements/rubin-prerelease.txt" >&2; exit 1; fi && if [ "$(echo $CUDA_VERSION | cut -d. -f1)" = "12" ]; then sed -i 's/^nvidia-cutlass-dsl\[cu13\]/nvidia-cutlass-dsl/' /tmp/requirements-cuda.txt; sed -i 's/^humming-kernels\[cu13\]/humming-kernels[cu12]/' /tmp/requirements-cuda.txt; fi && if [ "${INSTALL_RUBIN_PRERELEASE}" = "true" ]; then uv pip install --system --prerelease=allow --index-strategy unsafe-first-match -r /tmp/vllm-requirements/rubin-prerelease.txt && sed -Ei '/^[[:space:]]*(torch|torchvision|torchaudio)[[:space:]]*==/Id' /tmp/requirements-cuda.txt && uv pip install --system -r /tmp/requirements-cuda.txt && uv pip install --system --prerelease=allow --index-strategy unsafe-first-match -r /tmp/vllm-requirements/rubin-prerelease.txt; else uv pip install --system -r /tmp/requirements-cuda.txt --extra-index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.'); fi && rm /tmp/requirements-cuda.txt /tmp/common.txt # buildkit 3.42 GB
  59. 59 RUN |10 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl /bin/sh -c PIP_CACHE_DIR=/opt/pip/cache flashinfer download-kernels # buildkit 1.46 GB
  60. 60 ARG GDRCOPY_CUDA_VERSION=12.8 0 B
  61. 61 ARG GDRCOPY_OS_VERSION=Ubuntu24_04 0 B
  62. 62 ARG TARGETPLATFORM 0 B
  63. 63 COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit 940 Bytes
  64. 64 RUN |13 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 /bin/sh -c set -eux; case "${TARGETPLATFORM}" in linux/arm64) UUARCH="aarch64" ;; linux/amd64) UUARCH="x64" ;; *) echo "Unsupported TARGETPLATFORM: ${TARGETPLATFORM}" >&2; exit 1 ;; esac; /tmp/install_gdrcopy.sh "${GDRCOPY_OS_VERSION}" "${GDRCOPY_CUDA_VERSION}" "${UUARCH}" && rm /tmp/install_gdrcopy.sh # buildkit 428 kB
  65. 65 ARG TIMM_VERSION=>=1.0.17 0 B
  66. 66 ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 0 B
  67. 67 RUN |15 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/amd64 max_jobs=16 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 /bin/sh -c uv pip install --system accelerate 'modelscope<1.38' "timm${TIMM_VERSION}" "runai-model-streamer[s3,gcs,azure]${RUNAI_MODEL_STREAMER_VERSION}" # buildkit 60.1 MB
  68. 68 ARG PIP_INDEX_URL UV_INDEX_URL 0 B
  69. 69 ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL 0 B
  70. 70 ARG PYTORCH_CUDA_INDEX_BASE_URL 0 B
  71. 71 ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER 0 B
  72. 72 ARG PYTORCH_NIGHTLY 0 B
  73. 73 COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit 187 Bytes
  74. 74 COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit 230 Bytes
  75. 75 RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled PYTORCH_NIGHTLY= /bin/sh -c if [ "${PYTORCH_NIGHTLY}" = "1" ]; then echo "Installing torch nightly..." && uv pip install --system $(cat torch_lib_versions.txt | xargs) --pre --index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && echo "Installing vLLM..." && uv pip install --system dist/*.whl --verbose --extra-index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/nightly/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.'); else echo "Installing vLLM..." && uv pip install --system dist/*.whl --verbose --extra-index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.'); fi # buildkit 283 MB
  76. 76 RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled PYTORCH_NIGHTLY= /bin/sh -c ln -s "$(python3 -c 'import sysconfig; print(sysconfig.get_path("platlib"))')/vllm/vllm-rs" /usr/local/bin/vllm-rs && vllm-rs --version # buildkit 198 Bytes
  77. 77 RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled PYTORCH_NIGHTLY= /bin/sh -c . /etc/environment && uv pip list # buildkit 32 Bytes
  78. 78 ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64 0 B
  79. 79 COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit 248 Bytes
  80. 80 RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled PYTORCH_NIGHTLY= /bin/sh -c uv pip install --system ep_kernels/dist/*.whl --verbose --extra-index-url ${PYTORCH_CUDA_INDEX_BASE_URL}/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') # buildkit 12.7 MB
  81. 81 ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64 0 B
  82. 82 COPY examples examples # buildkit 311 kB
  83. 83 COPY benchmarks benchmarks # buildkit 304 kB
  84. 84 COPY ./vllm/collect_env.py . # buildkit 10.6 kB
  85. 85 ARG TARGETPLATFORM 0 B
  86. 86 ARG INSTALL_KV_CONNECTORS=false 0 B
  87. 87 ARG INSTALL_RUBIN_PRERELEASE 0 B
  88. 88 ARG TRITON_INSTALL_FROM_SOURCE_REPO 0 B
  89. 89 ARG CUDA_VERSION 0 B
  90. 90 ARG VLLM_BUILD_COMMIT 0 B
  91. 91 ARG VLLM_BUILD_PIPELINE 0 B
  92. 92 ARG VLLM_BUILD_URL 0 B
  93. 93 ARG VLLM_IMAGE_TAG 0 B
  94. 94 ARG PIP_INDEX_URL UV_INDEX_URL 0 B
  95. 95 ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL 0 B
  96. 96 ENV UV_HTTP_TIMEOUT=500 0 B
  97. 97 ARG torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 11.0 12.0 0 B
  98. 98 ENV TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 8.9 9.0 10.0 12.0 0 B
  99. 99 RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false TRITON_INSTALL_FROM_SOURCE_REPO= CUDA_VERSION=13.0.2 VLLM_BUILD_COMMIT=5df63a7203cdb9328d824a27b682e77c7b9928be VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/8622 VLLM_IMAGE_TAG=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 12.0 /bin/sh -c set -eu; CUDA_MAJOR="${CUDA_VERSION%%.*}"; CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-'); CUDA_HOME=/usr/local/cuda; BUILD_PKGS="libcusparse-dev-${CUDA_VERSION_DASH} libcublas-dev-${CUDA_VERSION_DASH} libcusolver-dev-${CUDA_VERSION_DASH}"; if [ "$INSTALL_KV_CONNECTORS" = "true" ]; then KV_REQUIREMENTS=/tmp/vllm-requirements/kv_connectors.txt; if [ "${CUDA_MAJOR}" = "12" ]; then KV_REQUIREMENTS=/tmp/vllm-requirements/kv_connectors_cu12.txt; fi; uv pip freeze --system > /tmp/kv-installed.txt; grep -E '^(torch|torchaudio|torchvision|triton|cuda-bindings|cuda-python|cuda-toolkit)==' /tmp/kv-installed.txt > /tmp/kv-constraints.txt; uv pip install --system -c /tmp/kv-constraints.txt -r "${KV_REQUIREMENTS}" --no-build || ( apt-get update -y && apt-get install -y --no-install-recommends --allow-change-held-packages ${BUILD_PKGS} && uv pip install --system -c /tmp/kv-constraints.txt -r "${KV_REQUIREMENTS}" --no-build-isolation && apt-get purge -y ${BUILD_PKGS} && rm -rf /var/lib/apt/lists/* ); MOONCAKE_VERSION=$(python3 -c "import importlib.metadata as m; print(m.version('mooncake-transfer-engine'))" 2>/dev/null || true) && if [ "${CUDA_MAJOR}" = "13" ] && [ -n "${MOONCAKE_VERSION}" ]; then uv pip uninstall --system mooncake-transfer-engine 2>/dev/null || true; uv pip install --system -c /tmp/kv-constraints.txt "mooncake-transfer-engine-cuda13==${MOONCAKE_VERSION}"; fi; rm -f /tmp/kv-installed.txt /tmp/kv-constraints.txt; fi # buildkit 305 MB
  100. 100 RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false TRITON_INSTALL_FROM_SOURCE_REPO= CUDA_VERSION=13.0.2 VLLM_BUILD_COMMIT=5df63a7203cdb9328d824a27b682e77c7b9928be VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/8622 VLLM_IMAGE_TAG=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 12.0 /bin/sh -c if [ "${INSTALL_RUBIN_PRERELEASE}" = "true" ]; then echo "Installing Rubin prerelease requirements..." && uv pip install --system --prerelease=allow --index-strategy unsafe-first-match -r /tmp/vllm-requirements/rubin-prerelease.txt; fi # buildkit 90 Bytes
  101. 101 COPY /tmp/triton_from_source_workspace/dist/wheel.sha256 /tmp/triton-source-wheel.sha256 # buildkit 216 Bytes
  102. 102 COPY tools/install_triton_from_source.sh /tmp/install_triton_from_source.sh # buildkit 849 Bytes
  103. 103 RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false TRITON_INSTALL_FROM_SOURCE_REPO= CUDA_VERSION=13.0.2 VLLM_BUILD_COMMIT=5df63a7203cdb9328d824a27b682e77c7b9928be VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/8622 VLLM_IMAGE_TAG=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 12.0 /bin/sh -c if [ -n "${TRITON_INSTALL_FROM_SOURCE_REPO}" ]; then CUDA_MAJOR_MINOR=$(echo "${CUDA_VERSION}" | cut -d. -f1,2); if [ "${INSTALL_RUBIN_PRERELEASE}" = "true" ]; then case "${CUDA_MAJOR_MINOR}" in 13.4|13.5) TRITON_SYSTEM_PTXAS_PATH=/usr/local/cuda/bin/ptxas TRITON_EXPECTED_PTXAS_VERSION="${CUDA_MAJOR_MINOR}" bash /tmp/install_triton_from_source.sh /tmp/triton-source-dist /tmp/triton-source-wheel.sha256 ;; *) echo "Unsupported Rubin CUDA version: ${CUDA_MAJOR_MINOR}" >&2; exit 1 ;; esac; else bash /tmp/install_triton_from_source.sh /tmp/triton-source-dist /tmp/triton-source-wheel.sha256; fi; fi && rm -f /tmp/install_triton_from_source.sh /tmp/triton-source-wheel.sha256 # buildkit 190 Bytes
  104. 104 ARG BUILD_NIXL 0 B
  105. 105 RUN |15 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false TRITON_INSTALL_FROM_SOURCE_REPO= CUDA_VERSION=13.0.2 VLLM_BUILD_COMMIT=5df63a7203cdb9328d824a27b682e77c7b9928be VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/8622 VLLM_IMAGE_TAG=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 12.0 BUILD_NIXL=false /bin/sh -c set -eu; if [ "${BUILD_NIXL}" = "true" ]; then ls /tmp/nixl-dist/*.whl; uv pip uninstall --system nixl nixl-cu12 nixl-cu13; uv pip install --system --no-deps /tmp/nixl-dist/*.whl; python3 -c "import glob, importlib.util as u, os, torch; spec = u.find_spec('nixl_ep_cu' + os.environ['CUDA_VERSION'].split('.')[0]); mm = ''.join(torch.__version__.split('.')[:2]); ext = glob.glob(os.path.join(spec.submodule_search_locations[0], 'nixl_ep_cpp_torch' + mm + '.*')) if spec else []; assert ext, 'NIXL EP has no build for torch ' + torch.__version__; print('NIXL EP extension:', ext[0])"; fi # buildkit 91 Bytes
  106. 106 ENV VLLM_USAGE_SOURCE=production-docker-image 0 B
  107. 107 ENV VLLM_BUILD_COMMIT=5df63a7203cdb9328d824a27b682e77c7b9928be VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/8622 VLLM_IMAGE_TAG=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be 0 B
  108. 108 LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=5df63a7203cdb9328d824a27b682e77c7b9928be org.opencontainers.image.version=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/8622 ai.vllm.build.commit=5df63a7203cdb9328d824a27b682e77c7b9928be ai.vllm.build.pipeline=019d130e-464e-4ff7-b84b-492992c0c06b ai.vllm.build.url=https://buildkite.com/vllm/release-v2/builds/8622 ai.vllm.image.tag=vllm/vllm-openai:v5df63a7203cdb9328d824a27b682e77c7b9928be 0 B
  109. 109 ENTRYPOINT ["vllm" "serve"] 0 B