Namespace
gpustack
Image / Tag
runner:cuda12.9-vllmv0.19.1
Content Digest
sha256:458f79651156abb2d6e653ff161bb72ade07ee8ddfa5bc685477441ad18a6b4d
Details
Created

2026-05-07 02:52:13 UTC

Size

9.2 GB

Content Digest
Labels
  • maintainer
    NVIDIA CORPORATION <cudatools@nvidia.com>
  • org.opencontainers.image.ref.name
    ubuntu
  • org.opencontainers.image.version
    24.04

Environment
CUDA_VERSION

12.9.1

DEBIAN_FRONTEND

noninteractive

LANG

en_US.UTF-8

LANGUAGE

en_US:en

LC_ALL

en_US.UTF-8

LD_LIBRARY_PATH

/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64

NVARCH

x86_64

NVIDIA_DRIVER_CAPABILITIES

compute,utility

NVIDIA_REQUIRE_CUDA

cuda>=12.9 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>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 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

NVIDIA_VISIBLE_DEVICES

all

NV_CUDA_CUDART_VERSION

12.9.79-1

PATH

/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

PYTHON_VERSION

3.12

RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES

1

RUNAI_STREAMER_LOG_LEVEL

INFO

RUNAI_STREAMER_LOG_TO_STDERR

1

RUNAI_STREAMER_MEMORY_LIMIT

0

SAFETENSORS_FAST_GPU

1

TORCH_CUDA_ARCH_LIST

8.7 8.9 9.0 10.0+PTX 12.0

UV_HTTP_TIMEOUT

500

UV_INDEX_STRATEGY

unsafe-best-match

UV_LINK_MODE

copy

UV_PRERELEASE

allow

UV_SKIP_WHEEL_FILENAME_CHECK

1

UV_SYSTEM_PYTHON

1

VLLM_ENABLE_CUDA_COMPATIBILITY

0

VLLM_TORCH_CUDA_VERSION

12.9.1

VLLM_TORCH_VERSION

2.10.0

VLLM_USAGE_SOURCE

production-docker-image

VLLM_VERSION

v0.19.1


Layers

[#000] sha256:e87500e698966458d9dfc34df84602985c9821f39666619792fe6282aa6df5d4 - 0.31% (29.2 MB)

[#001] sha256:644e9b20358325501941bab7efe2465969b1101fd546be263fc7c2d12d2d8c6c - 0.05% (4.34 MB)

[#002] sha256:02559cd4bc8db240554ff3a4e38df5909d4745846b3e3a4df865d102f0a6d0d9 - 1.05% (98.7 MB)

[#003] sha256:2cd52cbb1ebefad9b697ee1c60a7ba7ce8293404a1ad596fd257efdf0e9ec8d6 - 0.0% (182 Bytes)

[#004] sha256:6e8af4fd0a071982e528b634ba99dec2474c21147f99748be708f36e10e3f4c2 - 0.0% (6.72 KB)

[#005] sha256:24815637e8aa1fab3f297c9e4cdae723da8550ab28740ad8d6ea5002587067dd - 0.0% (104 Bytes)

[#006] sha256:1220f2950dbbdad8d52695a83c701935d0ebc6dcc642c64919bc73fe12d77fa7 - 0.0% (229 Bytes)

[#007] sha256:a8c386b98fc1d27bda873577b9d8cd35c5210027bffc4399f01a058ffd35614f - 2.36% (222 MB)

[#008] sha256:494aeb80506e48969dad4558c517cf48bb66849e4f82f71adea2ee302fa8145d - 14.79% (1.36 GB)

[#009] sha256:aa2e763adba6c22b5bb5247b5c0d04992b03e3e8f0d030cf2a54041397e0117a - 0.51% (48 MB)

[#010] sha256:0f6c809d0f7f11b206c0ec1a11d16705bdd2da986149969359afb95071d7de63 - 0.0% (1.52 KB)

[#011] sha256:b5f1b69b9459b45154fcb752f219349299dcf0f10bd5e3f310f5d355068874a2 - 0.0% (570 Bytes)

[#012] sha256:43c5621ba9357d40b05458e287ebc6d1acce4e9900a63672479e2a71d31a3245 - 53.49% (4.92 GB)

[#013] sha256:329eec3e98c9491634fe7e08fd2aa2a028930215ec6ef21242c64225fe3ad7d6 - 16.56% (1.52 GB)

[#014] sha256:ce5fbec7538f8082ce87cd2ac3bd5ce732713b6d653f94323ebd7b9b5875b030 - 0.0% (59.9 KB)

[#015] sha256:6edd8215f9aba7adcca79fb173a2f65350f75c28e989bb1ff0f0da6773fcdadc - 0.0% (897 Bytes)

[#016] sha256:ba7bc19b46fc7e325fcec7386d63baea3099cfb56cd7b7ef76db8b890f7248c1 - 0.01% (522 KB)

[#017] sha256:6f6574e13360e1cd9b76685391f103efd319b98992e00fd26418a3eb33fd8568 - 1.03% (96.8 MB)

[#018] sha256:3a83ef693b56aeeda9f7dde93bda5cce76c0e4a182be68cda30d17677f6f9c2a - 0.0% (199 Bytes)

[#019] sha256:b6ad2f034e283ea22a507847f95791193a249ba5d390dc4a34bd1812b0572e40 - 3.91% (368 MB)

[#020] sha256:e9d6b6af74f31934b60ae26be1d3161e07a9bdd2cd23db1d59348cf5ca88dd77 - 0.0% (121 Bytes)

[#021] sha256:78145a5fe0ce413b3a09ef143d998e6c900c526b1634fb39bfad17a0d752022f - 0.09% (8.16 MB)

[#022] sha256:0f506db5af23d1cd00e6804c05ca7a396c0647aad508d23c5dfef6f5610afab8 - 0.16% (15.4 MB)

[#023] sha256:696712e7960d22a243717e8dadad82dc81053baa9cdcf8fbebdb9c027f1636f4 - 0.0% (234 KB)

[#024] sha256:4d006ce99a4781e20227c8f27382a76de9f1ae99167b7f9c073e298400a1018d - 0.0% (217 KB)

[#025] sha256:d83b33e70ff76f51a37b8ab2bccd29c1c3a25f535149e19612b27e9abbca1c27 - 0.0% (8.18 KB)

[#026] sha256:9f365cf0d5bbf74519e698ad0554a3af0a52d73537b00c1e0368a595fc98e499 - 2.85% (268 MB)

[#027] sha256:62670c335c33023804defd03b45f2f13c4f045373196cbb4da6da51b2af9e750 - 1.04% (98.3 MB)

[#028] sha256:d9ef5ee90552db557714feb98fdcc01634723e6c8b1c7ebe6ecc74461642b185 - 0.0% (1.9 KB)

[#029] sha256:ad1a2f6ce0e1b6a97c21abbc68007172df6910047af99d05f8de2e035dbaae84 - 1.81% (170 MB)

[#030] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)

[#031] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)

[#032] sha256:1972b002a29cbde93ddc7dba65bd84c0539dddf35c075d70fc729e2cb79896c9 - 0.0% (85.8 KB)


History
2025-07-14 16:40:20 UTC

/bin/sh -c #(nop) ARG RELEASE

2025-07-14 16:40:20 UTC

/bin/sh -c #(nop) ARG LAUNCHPAD_BUILD_ARCH

2025-07-14 16:40:20 UTC

/bin/sh -c #(nop) LABEL org.opencontainers.image.ref.name=ubuntu

2025-07-14 16:40:20 UTC

/bin/sh -c #(nop) LABEL org.opencontainers.image.version=24.04

2025-07-14 16:40:22 UTC

/bin/sh -c #(nop) ADD file:b4619a63cd7829e1338ddaa4995ca17003002dd54b0dfd675a6f54a2b69151a6 in /

2025-07-14 16:40:23 UTC

/bin/sh -c #(nop) CMD ["/bin/bash"]

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

ENV NVARCH=x86_64

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

ENV NVIDIA_REQUIRE_CUDA=cuda>=12.9 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>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 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

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

ENV NV_CUDA_CUDART_VERSION=12.9.79-1

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

ARG TARGETARCH

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

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

2025-07-18 20:11:02 UTC (buildkit.dockerfile.v0)

ENV CUDA_VERSION=12.9.1

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

RUN |1 TARGETARCH=amd64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends cuda-cudart-12-9=${NV_CUDA_CUDART_VERSION} cuda-compat-12-9 && rm -rf /var/lib/apt/lists/* # buildkit

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

RUN |1 TARGETARCH=amd64 /bin/sh -c echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

COPY NGC-DL-CONTAINER-LICENSE / # buildkit

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

ENV NVIDIA_VISIBLE_DEVICES=all

2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)

ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

ARG CUDA_VERSION

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

ARG PYTHON_VERSION

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

ARG DEADSNAKES_MIRROR_URL

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

ARG DEADSNAKES_GPGKEY_URL

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

ARG GET_PIP_URL

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

ENV DEBIAN_FRONTEND=noninteractive

2026-04-18 01:01:33 UTC (buildkit.dockerfile.v0)

WORKDIR /vllm-workspace

2026-04-18 01:01:34 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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

2026-04-18 01:02:46 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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

2026-04-18 01:03:27 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-') && apt-get update -y && apt-get install -y --no-install-recommends cuda-nvcc-${CUDA_VERSION_DASH} cuda-cudart-${CUDA_VERSION_DASH} cuda-nvrtc-${CUDA_VERSION_DASH} cuda-cuobjdump-${CUDA_VERSION_DASH} libcurand-dev-${CUDA_VERSION_DASH} libcublas-${CUDA_VERSION_DASH} libnccl-dev && rm -rf /var/lib/apt/lists/* # buildkit

2026-04-18 01:03:28 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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

2026-04-18 01:03:28 UTC (buildkit.dockerfile.v0)

ENV UV_HTTP_TIMEOUT=500

2026-04-18 01:03:28 UTC (buildkit.dockerfile.v0)

ENV UV_INDEX_STRATEGY=unsafe-best-match

2026-04-18 01:03:28 UTC (buildkit.dockerfile.v0)

ENV UV_LINK_MODE=copy

2026-04-18 01:03:28 UTC (buildkit.dockerfile.v0)

ENV VLLM_ENABLE_CUDA_COMPATIBILITY=0

2026-04-18 01:03:28 UTC (buildkit.dockerfile.v0)

ARG PYTORCH_CUDA_INDEX_BASE_URL

2026-04-18 01:03:29 UTC (buildkit.dockerfile.v0)

COPY requirements/common.txt /tmp/common.txt # buildkit

2026-04-18 01:03:29 UTC (buildkit.dockerfile.v0)

COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit

2026-04-18 01:04:49 UTC (buildkit.dockerfile.v0)

RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 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 '.') && rm /tmp/requirements-cuda.txt /tmp/common.txt # buildkit

2026-04-18 01:04:49 UTC (buildkit.dockerfile.v0)

ARG FLASHINFER_VERSION=0.6.6

2026-04-18 01:05:55 UTC (buildkit.dockerfile.v0)

RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 /bin/sh -c uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} --extra-index-url https://flashinfer.ai/whl/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && flashinfer show-config # buildkit

2026-04-18 01:06:01 UTC (buildkit.dockerfile.v0)

RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 /bin/sh -c python3 <<'PYEOF' from flashinfer.jit import env as jit_env from flashinfer.jit.cubin_loader import download_trtllm_headers, get_cubin from flashinfer.artifacts import ArtifactPath, CheckSumHash download_trtllm_headers( 'bmm', jit_env.FLASHINFER_CUBIN_DIR / 'flashinfer' / 'trtllm' / 'batched_gemm' / 'trtllmGen_bmm_export', f'{ArtifactPath.TRTLLM_GEN_BMM}/include/trtllmGen_bmm_export', ArtifactPath.TRTLLM_GEN_BMM, get_cubin(f'{ArtifactPath.TRTLLM_GEN_BMM}/checksums.txt', CheckSumHash.TRTLLM_GEN_BMM), ) print('FlashInfer TRTLLM BMM headers downloaded successfully') PYEOF # buildkit

2026-04-18 01:06:01 UTC (buildkit.dockerfile.v0)

ARG GDRCOPY_CUDA_VERSION=12.8

2026-04-18 01:06:01 UTC (buildkit.dockerfile.v0)

ARG GDRCOPY_OS_VERSION=Ubuntu22_04

2026-04-18 01:06:01 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM

2026-04-18 01:06:01 UTC (buildkit.dockerfile.v0)

COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit

2026-04-18 01:06:06 UTC (buildkit.dockerfile.v0)

RUN |10 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 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

2026-04-18 01:06:06 UTC (buildkit.dockerfile.v0)

ARG BITSANDBYTES_VERSION_X86=0.46.1

2026-04-18 01:06:06 UTC (buildkit.dockerfile.v0)

ARG BITSANDBYTES_VERSION_ARM64=0.42.0

2026-04-18 01:06:06 UTC (buildkit.dockerfile.v0)

ARG TIMM_VERSION=>=1.0.17

2026-04-18 01:06:06 UTC (buildkit.dockerfile.v0)

ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.7

2026-04-18 01:06:09 UTC (buildkit.dockerfile.v0)

RUN |14 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 /bin/sh -c if [ "$TARGETPLATFORM" = "linux/arm64" ]; then BITSANDBYTES_VERSION="${BITSANDBYTES_VERSION_ARM64}"; else BITSANDBYTES_VERSION="${BITSANDBYTES_VERSION_X86}"; fi; uv pip install --system accelerate modelscope "bitsandbytes>=${BITSANDBYTES_VERSION}" "timm${TIMM_VERSION}" "runai-model-streamer[s3,gcs,azure]${RUNAI_MODEL_STREAMER_VERSION}" # buildkit

2026-04-18 01:06:09 UTC (buildkit.dockerfile.v0)

ARG PIP_INDEX_URL UV_INDEX_URL

2026-04-18 01:06:09 UTC (buildkit.dockerfile.v0)

ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL

2026-04-18 01:06:09 UTC (buildkit.dockerfile.v0)

ARG PYTORCH_CUDA_INDEX_BASE_URL

2026-04-18 01:06:09 UTC (buildkit.dockerfile.v0)

ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER

2026-04-18 01:06:09 UTC (buildkit.dockerfile.v0)

ARG PYTORCH_NIGHTLY

2026-04-18 01:06:10 UTC (buildkit.dockerfile.v0)

COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit

2026-04-18 01:18:00 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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

2026-04-18 01:18:01 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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

2026-04-18 01:18:02 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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 sh -c 'if ls /tmp/deepgemm/dist/*.whl >/dev/null 2>&1; then uv pip install --system /tmp/deepgemm/dist/*.whl; else echo "No DeepGEMM wheels to install; skipping."; fi' # buildkit

2026-04-18 01:18:02 UTC (buildkit.dockerfile.v0)

ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

COPY examples examples # buildkit

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

COPY benchmarks benchmarks # buildkit

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

COPY ./vllm/collect_env.py . # buildkit

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ARG INSTALL_KV_CONNECTORS=false

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ARG CUDA_VERSION

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ARG PIP_INDEX_URL UV_INDEX_URL

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ENV UV_HTTP_TIMEOUT=500

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ARG torch_cuda_arch_list=7.0 7.5 8.0 8.9 9.0 10.0 12.0

2026-04-18 01:18:03 UTC (buildkit.dockerfile.v0)

ENV TORCH_CUDA_ARCH_LIST=8.7 8.9 9.0 10.0+PTX 12.0

2026-04-18 01:18:10 UTC (buildkit.dockerfile.v0)

RUN |8 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=12.9.1 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=8.7 8.9 9.0 10.0+PTX 12.0 /bin/sh -c 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 [ "$INSTALL_KV_CONNECTORS" = "true" ]; then if [ "$CUDA_MAJOR" -ge 13 ]; then uv pip install --system nixl-cu13; fi; uv pip install --system -r /tmp/kv_connectors.txt --no-build || ( apt-get update -y && apt-get install -y --no-install-recommends ${BUILD_PKGS} && uv pip install --system -r /tmp/kv_connectors.txt --no-build-isolation && apt-get purge -y ${BUILD_PKGS} && rm -rf /var/lib/apt/lists/* ); fi # buildkit

2026-04-18 01:18:10 UTC (buildkit.dockerfile.v0)

ENV VLLM_USAGE_SOURCE=production-docker-image

2026-04-18 01:18:10 UTC (buildkit.dockerfile.v0)

ENTRYPOINT ["vllm" "serve"]

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

SHELL [/bin/bash -eo pipefail -c]

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM=linux/amd64

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG TARGETOS=linux

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG TARGETARCH=amd64

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ENV DEBIAN_FRONTEND=noninteractive LANG=en_US.UTF-8 LANGUAGE=en_US:en LC_ALL=en_US.UTF-8

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

RUN |3 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 /bin/bash -eo pipefail -c # Tools # Refresh apt-get update -y && apt-get install -y --no-install-recommends \ software-properties-common apt-transport-https \ ca-certificates gnupg2 lsb-release gnupg-agent \ && apt-get update -y # Install apt-get install -y --no-install-recommends \ ca-certificates build-essential binutils bash openssl \ curl wget aria2 \ git git-lfs \ unzip xz-utils \ tzdata locales \ iproute2 iputils-ping ifstat net-tools dnsutils pciutils ipmitool \ rdma-core rdmacm-utils infiniband-diags \ procps sysstat htop \ tini vim jq bc tree # Update locale localedef -i en_US -c -f UTF-8 -A /usr/share/locale/locale.alias en_US.UTF-8 # Update timezone rm -f /etc/localtime \ && ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime \ && echo "Asia/Shanghai" > /etc/timezone \ && dpkg-reconfigure --frontend noninteractive tzdata # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* \ && rm -rf /var/cache/apt # buildkit

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG PYTHON_VERSION=3.12

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ENV PYTHON_VERSION=3.12

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

RUN |4 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 /bin/bash -eo pipefail -c # Python # Update alternatives if [[ -f /etc/alternatives/python ]]; then update-alternatives --remove-all python; fi; update-alternatives --install /usr/bin/python python /usr/bin/python${PYTHON_VERSION} 1 # buildkit

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ENV UV_SYSTEM_PYTHON=1 UV_PRERELEASE=allow UV_SKIP_WHEEL_FILENAME_CHECK=1

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG VLLM_VERSION=v0.19.1

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG VLLM_TORCH_VERSION=2.10.0

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG VLLM_TORCH_CUDA_VERSION=12.9.1

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ENV VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

SHELL [/bin/bash -eo pipefail -c]

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM=linux/amd64

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG TARGETOS=linux

2026-05-07 02:51:45 UTC (buildkit.dockerfile.v0)

ARG TARGETARCH=amd64

2026-05-07 02:51:48 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Ray # Install Ray Client and Default RAY_VERSION=2.54.1 cat <<EOT >/tmp/requirements.txt ray[cgraph]==${RAY_VERSION} ray[client]==${RAY_VERSION} ray[default]==${RAY_VERSION} EOT uv pip install \ -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit

2026-05-07 02:52:12 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Omni if [[ ! -d /omni/workspace ]]; then echo "Skipping vLLM Omni installation for ${TARGETARCH}..." exit 0 fi # Install uv pip install --no-build-isolation \ /omni/workspace/*.whl # Dependencies cat <<EOT >/tmp/requirements.txt onnxruntime sox EOT uv pip install \ -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit

2026-05-07 02:52:12 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Postprocess # Review uv pip tree \ --package vllm \ --package vllm-omni \ --package flashinfer-python \ --package torch \ --package triton \ --package deep-gemm \ --package deep-ep \ --package lmcache \ --package mooncake-transfer-engine \ --package ray # buildkit

2026-05-07 02:52:13 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Patch tree -hs /workspace/patches pushd $(pip show vllm | grep Location: | cut -d" " -f 2) \ && patch -p1 < /workspace/patches/vllm/*.patch if pip show vllm_omni > /dev/null 2>&1; then \ pushd $(pip show vllm_omni | grep Location: | cut -d" " -f 2) \ && patch -p1 < /workspace/patches/vllm_omni/*.patch; \ fi # buildkit

2026-05-07 02:52:13 UTC (buildkit.dockerfile.v0)

ENV RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES=1 SAFETENSORS_FAST_GPU=1 RUNAI_STREAMER_MEMORY_LIMIT=0 RUNAI_STREAMER_LOG_TO_STDERR=1 RUNAI_STREAMER_LOG_LEVEL=INFO

2026-05-07 02:52:13 UTC (buildkit.dockerfile.v0)

WORKDIR /

2026-05-07 02:52:13 UTC (buildkit.dockerfile.v0)

ENTRYPOINT ["tini" "--"]

Details
Created

2026-05-07 02:51:50 UTC

Size

10.1 GB

Content Digest
Labels
  • maintainer
    NVIDIA CORPORATION <cudatools@nvidia.com>
  • org.opencontainers.image.ref.name
    ubuntu
  • org.opencontainers.image.version
    24.04

Environment
CUDA_VERSION

12.9.1

DEBIAN_FRONTEND

noninteractive

LANG

en_US.UTF-8

LANGUAGE

en_US:en

LC_ALL

en_US.UTF-8

LD_LIBRARY_PATH

/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64

NVARCH

sbsa

NVIDIA_DRIVER_CAPABILITIES

compute,utility

NVIDIA_REQUIRE_CUDA

cuda>=12.9 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>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 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

NVIDIA_VISIBLE_DEVICES

all

NV_CUDA_CUDART_VERSION

12.9.79-1

PATH

/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

PYTHON_VERSION

3.12

RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES

1

RUNAI_STREAMER_LOG_LEVEL

INFO

RUNAI_STREAMER_LOG_TO_STDERR

1

RUNAI_STREAMER_MEMORY_LIMIT

0

SAFETENSORS_FAST_GPU

1

TORCH_CUDA_ARCH_LIST

8.7 8.9 9.0 10.0+PTX 12.0

UV_HTTP_TIMEOUT

500

UV_INDEX_STRATEGY

unsafe-best-match

UV_LINK_MODE

copy

UV_PRERELEASE

allow

UV_SKIP_WHEEL_FILENAME_CHECK

1

UV_SYSTEM_PYTHON

1

VLLM_ENABLE_CUDA_COMPATIBILITY

0

VLLM_TORCH_CUDA_VERSION

12.9.1

VLLM_TORCH_VERSION

2.10.0

VLLM_USAGE_SOURCE

production-docker-image

VLLM_VERSION

v0.19.1


Layers

[#000] sha256:1fe4490d3380a3fec7f0b5cfd536dbdeed91ff50c5fbd11eba4ec838d689855e - 0.28% (28.6 MB)

[#001] sha256:ecbca12713d0ee06710d449b9cecb86e2b3be1c32e2f7e6fd6ff3dea9219cf03 - 0.04% (4.32 MB)

[#002] sha256:ed5ca6b9e891c487cf1c2321e0b3e28f63b7ffdc2ee7c75592578bfcdebac740 - 0.9% (93.1 MB)

[#003] sha256:3f0047ea99ac7fb8902bd5e3aeb361d80ffa04c40b26b7d3935b6709c3c66f0c - 0.0% (182 Bytes)

[#004] sha256:6e8af4fd0a071982e528b634ba99dec2474c21147f99748be708f36e10e3f4c2 - 0.0% (6.72 KB)

[#005] sha256:e37f410a009119a99f1b65e59210535ae654724a85d0f8522682d4cf1198e4e4 - 0.0% (104 Bytes)

[#006] sha256:a55af85b35dbf446e0b9ab15ba01f6d09102067276bb6afd7ab4165a15ae2837 - 0.0% (229 Bytes)

[#007] sha256:7d0aac94166bf1f9857f5af289199c5b60f48dfc76583f372f479075859b0a31 - 2.09% (216 MB)

[#008] sha256:b6f93fa9c576f7f66793981bd0419df016fbc7fc507e0e3889808bc2c2d6972f - 12.96% (1.31 GB)

[#009] sha256:a187b153ca891b768fe375f5629542c8b0568e4e4fef2c72b8e877d0654602bc - 0.44% (45.9 MB)

[#010] sha256:8d1e1afd063adf17a9fb89e8b011beb6eb1d34ec65c02f1ecd0f27e3bdfc6515 - 0.0% (1.52 KB)

[#011] sha256:171c337ba62a8202293431725441cd394de074c43a5989844b5f1752a8519d3c - 0.0% (573 Bytes)

[#012] sha256:bf77733bd93dd1cf7755944c56d6e508f67fca2cfc3057bcd3120725e16f75cd - 44.3% (4.46 GB)

[#013] sha256:38eb2dbac327d3154aebdf7d2e2074962f810e6c1ade4290f45c611a85dedfb9 - 15.0% (1.51 GB)

[#014] sha256:e4e4f70cb49300fe0f6e9be09830cd08b5bb13b479d5725aac96b6053d43fdfe - 0.0% (59.9 KB)

[#015] sha256:fcbb782765f5cdd16b0d1bd80ef4333f4ae47aa577ce2b6bfb75d6e66a5993c1 - 0.0% (897 Bytes)

[#016] sha256:cf1e07703ada66ad400d7d1712e1c33156ef152e5897a84aabcfabb145cbdccf - 0.0% (517 KB)

[#017] sha256:96658202244017b5ddf8f5b00e5b07ba177bdcc76825f24caf7908c583347046 - 3.31% (342 MB)

[#018] sha256:a402474e5a15303222620e0ecafd7cb72326511ee07f007d9f355d4f207d5d48 - 0.0% (198 Bytes)

[#019] sha256:a0d0f5a419ff4ace9804677d76c702715c0c4e582b6963b1e5eeed6f7e6bb47c - 6.19% (639 MB)

[#020] sha256:36a58d7c04f11cc7fb71d741881a3813504f5c232329ff562d65c1a32fc54eff - 0.0% (122 Bytes)

[#021] sha256:983d690defc3a165c4bdf2649fcaaceb02a24cc7172b4c3e8bb83802ac4de759 - 0.08% (7.92 MB)

[#022] sha256:ff70a7aa551b24f59d1a01a50272bc5b22ca91288506f8fe78e9dc347249becc - 0.15% (15.2 MB)

[#023] sha256:7584f4187d029a455dc6f801c13d134165665ff7b02d86fa9dd9c7a93d06ef10 - 0.0% (234 KB)

[#024] sha256:45b5752f6ccc497199d822172da84e722c69fea749e7b0ad770223891c9746fb - 0.0% (217 KB)

[#025] sha256:e0f5c4706c20f4bb4845d0e90397a53b815847770b55409cce28d843a73dc9d1 - 0.0% (8.18 KB)

[#026] sha256:99d7d801117541b869ac81ba00eab0d048be33806e5ef860674ba2df9121a131 - 11.64% (1.17 GB)

[#027] sha256:b796c062af6576de9486d243b4684ba319983f3d4c0b48e637c2e167c7834449 - 0.98% (101 MB)

[#028] sha256:e7494ac9535ce6c2d9ee5008893a0a208de17085889fede9dca7d6e95fe4ea08 - 0.0% (1.89 KB)

[#029] sha256:3c9ab21c8a06d603f87045429633fd396922f7aca9bcfdf7587a385133ba3aff - 1.64% (169 MB)

[#030] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)

[#031] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)

[#032] sha256:1c1516dc35527082946a8fe587b5084e4588e2d8ab2dcdba63fda39beba12477 - 0.0% (85.8 KB)


History
2025-07-14 16:42:11 UTC

/bin/sh -c #(nop) ARG RELEASE

2025-07-14 16:42:11 UTC

/bin/sh -c #(nop) ARG LAUNCHPAD_BUILD_ARCH

2025-07-14 16:42:11 UTC

/bin/sh -c #(nop) LABEL org.opencontainers.image.ref.name=ubuntu

2025-07-14 16:42:12 UTC

/bin/sh -c #(nop) LABEL org.opencontainers.image.version=24.04

2025-07-14 16:42:14 UTC

/bin/sh -c #(nop) ADD file:98827745561bc42c12745b10037415621cf3b8ec8497a5953995b5fc7969bf17 in /

2025-07-14 16:42:15 UTC

/bin/sh -c #(nop) CMD ["/bin/bash"]

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

ENV NVARCH=sbsa

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

ENV NVIDIA_REQUIRE_CUDA=cuda>=12.9 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>=560,driver<561 brand=grid,driver>=560,driver<561 brand=tesla,driver>=560,driver<561 brand=nvidia,driver>=560,driver<561 brand=quadro,driver>=560,driver<561 brand=quadrortx,driver>=560,driver<561 brand=nvidiartx,driver>=560,driver<561 brand=vapps,driver>=560,driver<561 brand=vpc,driver>=560,driver<561 brand=vcs,driver>=560,driver<561 brand=vws,driver>=560,driver<561 brand=cloudgaming,driver>=560,driver<561 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

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

ENV NV_CUDA_CUDART_VERSION=12.9.79-1

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

ARG TARGETARCH

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

RUN |1 TARGETARCH=arm64 /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

2025-07-18 20:17:43 UTC (buildkit.dockerfile.v0)

ENV CUDA_VERSION=12.9.1

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

RUN |1 TARGETARCH=arm64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends cuda-cudart-12-9=${NV_CUDA_CUDART_VERSION} cuda-compat-12-9 && rm -rf /var/lib/apt/lists/* # buildkit

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

RUN |1 TARGETARCH=arm64 /bin/sh -c echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

COPY NGC-DL-CONTAINER-LICENSE / # buildkit

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

ENV NVIDIA_VISIBLE_DEVICES=all

2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)

ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

ARG CUDA_VERSION

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

ARG PYTHON_VERSION

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

ARG DEADSNAKES_MIRROR_URL

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

ARG DEADSNAKES_GPGKEY_URL

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

ARG GET_PIP_URL

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

ENV DEBIAN_FRONTEND=noninteractive

2026-04-18 01:00:54 UTC (buildkit.dockerfile.v0)

WORKDIR /vllm-workspace

2026-04-18 01:00:55 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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

2026-04-18 01:01:57 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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

2026-04-18 01:02:29 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-') && apt-get update -y && apt-get install -y --no-install-recommends cuda-nvcc-${CUDA_VERSION_DASH} cuda-cudart-${CUDA_VERSION_DASH} cuda-nvrtc-${CUDA_VERSION_DASH} cuda-cuobjdump-${CUDA_VERSION_DASH} libcurand-dev-${CUDA_VERSION_DASH} libcublas-${CUDA_VERSION_DASH} libnccl-dev && rm -rf /var/lib/apt/lists/* # buildkit

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

ENV UV_HTTP_TIMEOUT=500

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

ENV UV_INDEX_STRATEGY=unsafe-best-match

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

ENV UV_LINK_MODE=copy

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

ENV VLLM_ENABLE_CUDA_COMPATIBILITY=0

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

ARG PYTORCH_CUDA_INDEX_BASE_URL

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

COPY requirements/common.txt /tmp/common.txt # buildkit

2026-04-18 01:02:32 UTC (buildkit.dockerfile.v0)

COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit

2026-04-18 01:03:33 UTC (buildkit.dockerfile.v0)

RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 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 '.') && rm /tmp/requirements-cuda.txt /tmp/common.txt # buildkit

2026-04-18 01:03:33 UTC (buildkit.dockerfile.v0)

ARG FLASHINFER_VERSION=0.6.6

2026-04-18 01:04:24 UTC (buildkit.dockerfile.v0)

RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 /bin/sh -c uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} --extra-index-url https://flashinfer.ai/whl/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && flashinfer show-config # buildkit

2026-04-18 01:04:30 UTC (buildkit.dockerfile.v0)

RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 /bin/sh -c python3 <<'PYEOF' from flashinfer.jit import env as jit_env from flashinfer.jit.cubin_loader import download_trtllm_headers, get_cubin from flashinfer.artifacts import ArtifactPath, CheckSumHash download_trtllm_headers( 'bmm', jit_env.FLASHINFER_CUBIN_DIR / 'flashinfer' / 'trtllm' / 'batched_gemm' / 'trtllmGen_bmm_export', f'{ArtifactPath.TRTLLM_GEN_BMM}/include/trtllmGen_bmm_export', ArtifactPath.TRTLLM_GEN_BMM, get_cubin(f'{ArtifactPath.TRTLLM_GEN_BMM}/checksums.txt', CheckSumHash.TRTLLM_GEN_BMM), ) print('FlashInfer TRTLLM BMM headers downloaded successfully') PYEOF # buildkit

2026-04-18 01:04:30 UTC (buildkit.dockerfile.v0)

ARG GDRCOPY_CUDA_VERSION=12.8

2026-04-18 01:04:30 UTC (buildkit.dockerfile.v0)

ARG GDRCOPY_OS_VERSION=Ubuntu22_04

2026-04-18 01:04:30 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM

2026-04-18 01:04:30 UTC (buildkit.dockerfile.v0)

COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit

2026-04-18 01:04:35 UTC (buildkit.dockerfile.v0)

RUN |10 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 /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

2026-04-18 01:04:35 UTC (buildkit.dockerfile.v0)

ARG BITSANDBYTES_VERSION_X86=0.46.1

2026-04-18 01:04:35 UTC (buildkit.dockerfile.v0)

ARG BITSANDBYTES_VERSION_ARM64=0.42.0

2026-04-18 01:04:35 UTC (buildkit.dockerfile.v0)

ARG TIMM_VERSION=>=1.0.17

2026-04-18 01:04:35 UTC (buildkit.dockerfile.v0)

ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.7

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

RUN |14 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 /bin/sh -c if [ "$TARGETPLATFORM" = "linux/arm64" ]; then BITSANDBYTES_VERSION="${BITSANDBYTES_VERSION_ARM64}"; else BITSANDBYTES_VERSION="${BITSANDBYTES_VERSION_X86}"; fi; uv pip install --system accelerate modelscope "bitsandbytes>=${BITSANDBYTES_VERSION}" "timm${TIMM_VERSION}" "runai-model-streamer[s3,gcs,azure]${RUNAI_MODEL_STREAMER_VERSION}" # buildkit

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

ARG PIP_INDEX_URL UV_INDEX_URL

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

ARG PYTORCH_CUDA_INDEX_BASE_URL

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

ARG PYTORCH_NIGHTLY

2026-04-18 01:04:41 UTC (buildkit.dockerfile.v0)

COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit

2026-04-18 01:18:55 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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

2026-04-18 01:18:55 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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

2026-04-18 01:18:56 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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 sh -c 'if ls /tmp/deepgemm/dist/*.whl >/dev/null 2>&1; then uv pip install --system /tmp/deepgemm/dist/*.whl; else echo "No DeepGEMM wheels to install; skipping."; fi' # buildkit

2026-04-18 01:18:56 UTC (buildkit.dockerfile.v0)

ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64

2026-04-18 01:18:57 UTC (buildkit.dockerfile.v0)

RUN |22 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 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 FLASHINFER_VERSION=0.6.6 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 BITSANDBYTES_VERSION_X86=0.46.1 BITSANDBYTES_VERSION_ARM64=0.42.0 TIMM_VERSION=>=1.0.17 RUNAI_MODEL_STREAMER_VERSION=>=0.15.7 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

2026-04-18 01:18:57 UTC (buildkit.dockerfile.v0)

ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64

2026-04-18 01:18:57 UTC (buildkit.dockerfile.v0)

COPY examples examples # buildkit

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

COPY benchmarks benchmarks # buildkit

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

COPY ./vllm/collect_env.py . # buildkit

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ARG INSTALL_KV_CONNECTORS=false

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ARG CUDA_VERSION

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ARG PIP_INDEX_URL UV_INDEX_URL

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ENV UV_HTTP_TIMEOUT=500

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ARG torch_cuda_arch_list=7.0 7.5 8.0 8.9 9.0 10.0 12.0

2026-04-18 01:18:58 UTC (buildkit.dockerfile.v0)

ENV TORCH_CUDA_ARCH_LIST=8.7 8.9 9.0 10.0+PTX 12.0

2026-04-18 01:24:47 UTC (buildkit.dockerfile.v0)

RUN |8 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=12.9.1 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=8.7 8.9 9.0 10.0+PTX 12.0 /bin/sh -c 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 [ "$INSTALL_KV_CONNECTORS" = "true" ]; then if [ "$CUDA_MAJOR" -ge 13 ]; then uv pip install --system nixl-cu13; fi; uv pip install --system -r /tmp/kv_connectors.txt --no-build || ( apt-get update -y && apt-get install -y --no-install-recommends ${BUILD_PKGS} && uv pip install --system -r /tmp/kv_connectors.txt --no-build-isolation && apt-get purge -y ${BUILD_PKGS} && rm -rf /var/lib/apt/lists/* ); fi # buildkit

2026-04-18 01:24:47 UTC (buildkit.dockerfile.v0)

ENV VLLM_USAGE_SOURCE=production-docker-image

2026-04-18 01:24:47 UTC (buildkit.dockerfile.v0)

ENTRYPOINT ["vllm" "serve"]

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

SHELL [/bin/bash -eo pipefail -c]

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM=linux/arm64

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG TARGETOS=linux

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG TARGETARCH=arm64

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ENV DEBIAN_FRONTEND=noninteractive LANG=en_US.UTF-8 LANGUAGE=en_US:en LC_ALL=en_US.UTF-8

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

RUN |3 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 /bin/bash -eo pipefail -c # Tools # Refresh apt-get update -y && apt-get install -y --no-install-recommends \ software-properties-common apt-transport-https \ ca-certificates gnupg2 lsb-release gnupg-agent \ && apt-get update -y # Install apt-get install -y --no-install-recommends \ ca-certificates build-essential binutils bash openssl \ curl wget aria2 \ git git-lfs \ unzip xz-utils \ tzdata locales \ iproute2 iputils-ping ifstat net-tools dnsutils pciutils ipmitool \ rdma-core rdmacm-utils infiniband-diags \ procps sysstat htop \ tini vim jq bc tree # Update locale localedef -i en_US -c -f UTF-8 -A /usr/share/locale/locale.alias en_US.UTF-8 # Update timezone rm -f /etc/localtime \ && ln -sf /usr/share/zoneinfo/Asia/Shanghai /etc/localtime \ && echo "Asia/Shanghai" > /etc/timezone \ && dpkg-reconfigure --frontend noninteractive tzdata # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* \ && rm -rf /var/cache/apt # buildkit

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG PYTHON_VERSION=3.12

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ENV PYTHON_VERSION=3.12

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

RUN |4 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 /bin/bash -eo pipefail -c # Python # Update alternatives if [[ -f /etc/alternatives/python ]]; then update-alternatives --remove-all python; fi; update-alternatives --install /usr/bin/python python /usr/bin/python${PYTHON_VERSION} 1 # buildkit

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ENV UV_SYSTEM_PYTHON=1 UV_PRERELEASE=allow UV_SKIP_WHEEL_FILENAME_CHECK=1

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG VLLM_VERSION=v0.19.1

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG VLLM_TORCH_VERSION=2.10.0

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG VLLM_TORCH_CUDA_VERSION=12.9.1

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ENV VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

SHELL [/bin/bash -eo pipefail -c]

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG TARGETPLATFORM=linux/arm64

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG TARGETOS=linux

2026-05-07 02:30:42 UTC (buildkit.dockerfile.v0)

ARG TARGETARCH=arm64

2026-05-07 02:30:45 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Ray # Install Ray Client and Default RAY_VERSION=2.54.1 cat <<EOT >/tmp/requirements.txt ray[cgraph]==${RAY_VERSION} ray[client]==${RAY_VERSION} ray[default]==${RAY_VERSION} EOT uv pip install \ -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit

2026-05-07 02:31:04 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Omni if [[ ! -d /omni/workspace ]]; then echo "Skipping vLLM Omni installation for ${TARGETARCH}..." exit 0 fi # Install uv pip install --no-build-isolation \ /omni/workspace/*.whl # Dependencies cat <<EOT >/tmp/requirements.txt onnxruntime sox EOT uv pip install \ -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit

2026-05-07 02:31:04 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Postprocess # Review uv pip tree \ --package vllm \ --package vllm-omni \ --package flashinfer-python \ --package torch \ --package triton \ --package deep-gemm \ --package deep-ep \ --package lmcache \ --package mooncake-transfer-engine \ --package ray # buildkit

2026-05-07 02:51:50 UTC (buildkit.dockerfile.v0)

RUN |7 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 VLLM_VERSION=v0.19.1 VLLM_TORCH_VERSION=2.10.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Patch tree -hs /workspace/patches pushd $(pip show vllm | grep Location: | cut -d" " -f 2) \ && patch -p1 < /workspace/patches/vllm/*.patch if pip show vllm_omni > /dev/null 2>&1; then \ pushd $(pip show vllm_omni | grep Location: | cut -d" " -f 2) \ && patch -p1 < /workspace/patches/vllm_omni/*.patch; \ fi # buildkit

2026-05-07 02:51:50 UTC (buildkit.dockerfile.v0)

ENV RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES=1 SAFETENSORS_FAST_GPU=1 RUNAI_STREAMER_MEMORY_LIMIT=0 RUNAI_STREAMER_LOG_TO_STDERR=1 RUNAI_STREAMER_LOG_LEVEL=INFO

2026-05-07 02:51:50 UTC (buildkit.dockerfile.v0)

WORKDIR /

2026-05-07 02:51:50 UTC (buildkit.dockerfile.v0)

ENTRYPOINT ["tini" "--"]

Danger Zone
Delete Tag

Please be careful as this will not just delete the reference but also the actual content!

For example when you have latest and v1.2.3 both pointing to the same image
the deletion of latest will also permanently remove v1.2.3.

Delete