2026-06-05 14:06:40 UTC
10.2 GB
/usr/local/cuda
CUDA_VERSION12.9.1
DEBIAN_FRONTENDnoninteractive
LANGen_US.UTF-8
LANGUAGEen_US:en
LC_ALLen_US.UTF-8
LD_LIBRARY_PATH/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64
NVARCHx86_64
NVIDIA_DRIVER_CAPABILITIEScompute,utility
NVIDIA_REQUIRE_CUDAcuda>=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_DEVICESall
NV_CUDA_CUDART_VERSION12.9.79-1
PATH/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
PIP_DISABLE_PIP_VERSION_CHECK1
PIP_NO_CACHE_DIR1
PIP_ROOT_USER_ACTIONignore
PYTHON_VERSION3.12
RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES1
RUNAI_STREAMER_LOG_LEVELINFO
RUNAI_STREAMER_LOG_TO_STDERR1
RUNAI_STREAMER_MEMORY_LIMIT0
SAFETENSORS_FAST_GPU1
TORCH_CUDA_ARCH_LIST7.5 8.0 8.6 8.9 9.0 10.0 12.0
UV_HTTP_TIMEOUT500
UV_INDEX_STRATEGYunsafe-best-match
UV_LINK_MODEcopy
UV_NO_CACHE1
UV_PRERELEASEallow
UV_SKIP_WHEEL_FILENAME_CHECK1
UV_SYSTEM_PYTHON1
VLLM_ENABLE_CUDA_COMPATIBILITY0
VLLM_TORCH_CUDA_VERSION12.9.1
VLLM_TORCH_VERSION2.11.0
VLLM_USAGE_SOURCEproduction-docker-image
VLLM_VERSION0.20.2
[#000] sha256:e87500e698966458d9dfc34df84602985c9821f39666619792fe6282aa6df5d4 - 0.28% (29.2 MB)
[#001] sha256:644e9b20358325501941bab7efe2465969b1101fd546be263fc7c2d12d2d8c6c - 0.04% (4.34 MB)
[#002] sha256:02559cd4bc8db240554ff3a4e38df5909d4745846b3e3a4df865d102f0a6d0d9 - 0.94% (98.7 MB)
[#003] sha256:2cd52cbb1ebefad9b697ee1c60a7ba7ce8293404a1ad596fd257efdf0e9ec8d6 - 0.0% (182 Bytes)
[#004] sha256:6e8af4fd0a071982e528b634ba99dec2474c21147f99748be708f36e10e3f4c2 - 0.0% (6.72 KB)
[#005] sha256:13b966a991b62495f02b3333129cf66e548070f0ea6d8ed6bf8ba75efbe27830 - 0.0% (104 Bytes)
[#006] sha256:0bde045cc2cba19ad434681bce0b546d48cb7031d9ac8d517cb965af2f38e1fb - 0.0% (231 Bytes)
[#007] sha256:6e4a3d45453e17b361c19cf60b90d462c182c5a8ea501d374edbc308bcd0a40b - 3.72% (390 MB)
[#008] sha256:7b81dbb51fb31779e346425e9a923eb22313c76c081959cc76376904e81e8f1a - 14.11% (1.44 GB)
[#009] sha256:20de5ffd1f5a0e9feafdf58d5618175bdce0d544dd14cdb95f6a7a614a6c83c3 - 0.45% (47.5 MB)
[#010] sha256:91c207a9f785df659d1de4ff5f92e19adbbe55797ed91bad0e1e1a7956e82bdf - 0.0% (1.52 KB)
[#011] sha256:ea31fd39c18f4945c257c799b957abf2d39e0b8f885e2fbbecb3e55dbc04acdf - 0.0% (628 Bytes)
[#012] sha256:a64476f42174c22b8aac95c3e65e8aba42ec10d4a2c768909efe07191bd9bfff - 47.41% (4.85 GB)
[#013] sha256:6e1d9ee69612385679b586703f782fc8d939d55ceae8c5ff86ee2a332a752419 - 20.19% (2.06 GB)
[#014] sha256:fe2210b539614310232bfbb0661e4f7d449c7888ee72624129a2caea985b65ea - 0.0% (897 Bytes)
[#015] sha256:d0455538e0a946e254613d3b21db207809564abb99c28e7c89812850288315cf - 0.0% (492 KB)
[#016] sha256:7c1a3dde688eedb1d4e106beb70eec75b0167f2324c672aa8daf3b10fcd52f8d - 1.1% (115 MB)
[#017] sha256:8cdbd454c034103c6a71e880c25689eda5e168962c1227b65b4a3007cd0e8fa2 - 0.0% (200 Bytes)
[#018] sha256:5e0b7594508ce6eaca2d6586760e8a615479735736cd4bdb740dbcc6cd3a5358 - 4.12% (431 MB)
[#019] sha256:689c76db7ee3b2729e956637387d40cb55efa446d50140c143c7a3a80a5593af - 0.0% (123 Bytes)
[#020] sha256:21243f06ddf78ffec8798762152cc65aa5f28b11bb45d3ae5485c7f1afaf3bb8 - 0.15% (15.4 MB)
[#021] sha256:89149303322126cba98f95fec0002e3d08c515e14aeebb337af13579399c035f - 0.0% (241 KB)
[#022] sha256:870c7866d1daccb1f42badea257ce1a3bebb85d20d05ea4914fb251d7557757b - 0.0% (224 KB)
[#023] sha256:a03096b098bae39776e45a04df5625792eb4de1d3b860e207c6e8e93d2f2028c - 0.0% (9.62 KB)
[#024] sha256:e360644ae9913484def1bdd8144d6c3fbe1781f4b5ade7cfc7d5cba0e7561945 - 2.9% (303 MB)
[#025] sha256:e1a15c76b34ac016b36521c206c2685b34f3f70b39aa1d3ce6127414aab68fa0 - 1.19% (124 MB)
[#026] sha256:dd3ce9482c6731e4102a7ef29ed1cbf49e58df8f2492ac578a24af2c40dcfc1e - 0.0% (1.85 KB)
[#027] sha256:9930068f17aa062e6862b193bd5b85178d05b50a0b2155f51a7bc834077895dc - 0.3% (31.1 MB)
[#028] sha256:263a3923d4878dbc3a303fe01c1183e711b4ebf464d29969538ec6380fe54f9d - 0.78% (81.4 MB)
[#029] sha256:9213064f560eef24014062b5109df27f72864691a979351a3bc3df68c15dd530 - 1.46% (153 MB)
[#030] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#031] sha256:fcb92d0d1faa3457b0479b80f666ef8908ddf6d5b1c8f1423a29997634ede325 - 0.0% (6.2 KB)
[#032] sha256:1f04e0ab5563918579179fe05932ef2bd6b3c39ebdfc3d2773c4c9ce788c1988 - 0.09% (9.56 MB)
[#033] sha256:6efc206f5316b2ef270412ebcb9f1aa9852a93f11dcdc843734b4f0f73dd0ea2 - 0.0% (4.04 KB)
[#034] sha256:1a199475fc159530e9bc4fd8b4fab21b79a0716bda935a39cea44571ee587c70 - 0.75% (79 MB)
/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=${NV_CUDA_CUDART_VERSION} cuda-compat-12-9 && rm -rf /var/lib/apt/lists/* # buildkit
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=amd64 /bin/sh -c echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)COPY NGC-DL-CONTAINER-LICENSE / # buildkit
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV NVIDIA_VISIBLE_DEVICES=all
2025-07-18 20:11:18 UTC (buildkit.dockerfile.v0)ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)ARG PYTHON_VERSION
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)ARG DEADSNAKES_MIRROR_URL
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)ARG DEADSNAKES_GPGKEY_URL
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)ARG GET_PIP_URL
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)ENV DEBIAN_FRONTEND=noninteractive
2026-05-08 20:10:47 UTC (buildkit.dockerfile.v0)WORKDIR /vllm-workspace
2026-05-08 20:10:48 UTC (buildkit.dockerfile.v0)RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c PYTHON_VERSION_STR=$(echo ${PYTHON_VERSION} | sed 's/\.//g') && echo "export PYTHON_VERSION_STR=${PYTHON_VERSION_STR}" >> /etc/environment # buildkit
2026-05-08 20:14:10 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 curl sudo ffmpeg libsm6 libxext6 libgl1 libibverbs-dev build-essential libssl-dev libffi-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev libncurses-dev liblzma-dev libgdbm-dev uuid-dev tk-dev && PYTHON_MAJOR_MINOR=${PYTHON_VERSION} && PYTHON_FULL_VERSION=$(curl -s https://www.python.org/ftp/python/ | grep -oE "${PYTHON_MAJOR_MINOR}\.[0-9]+" | sort -t. -k3 -n | uniq | tail -1) && echo "Building Python ${PYTHON_FULL_VERSION} from source..." && curl -fSL https://www.python.org/ftp/python/${PYTHON_FULL_VERSION}/Python-${PYTHON_FULL_VERSION}.tgz -o /tmp/python.tgz && tar -xzf /tmp/python.tgz -C /tmp && cd /tmp/Python-${PYTHON_FULL_VERSION} && ./configure --enable-optimizations --with-ensurepip=install --prefix=/usr/local && make -j$(nproc) && make install && cd / && rm -rf /tmp/python.tgz /tmp/Python-${PYTHON_FULL_VERSION} && ln -sf /usr/local/bin/python${PYTHON_MAJOR_MINOR} /usr/bin/python3 && ln -sf /usr/local/bin/python${PYTHON_MAJOR_MINOR}-config /usr/bin/python3-config && ln -sf /usr/local/bin/pip${PYTHON_MAJOR_MINOR} /usr/bin/pip && rm -rf /var/lib/apt/lists/* && python3 --version && python3 -m pip --version # buildkit
2026-05-08 20:14:41 UTC (buildkit.dockerfile.v0)RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-') && CUDA_VERSION_SHORT=$(echo $CUDA_VERSION | cut -d. -f1,2) && apt-get update -y && apt-get install -y --no-install-recommends --allow-change-held-packages cuda-nvcc-${CUDA_VERSION_DASH} cuda-cudart-${CUDA_VERSION_DASH} cuda-nvrtc-${CUDA_VERSION_DASH} cuda-cuobjdump-${CUDA_VERSION_DASH} libcurand-dev-${CUDA_VERSION_DASH} libcublas-${CUDA_VERSION_DASH} libnuma-dev numactl && NCCL_VER=$(apt-cache madison libnccl-dev | grep "+cuda${CUDA_VERSION_SHORT}" | head -1 | awk -F'|' '{gsub(/^ +| +$/, "", $2); print $2}') && apt-get install -y --no-install-recommends --allow-change-held-packages libnccl-dev=${NCCL_VER} libnccl2=${NCCL_VER} && rm -rf /var/lib/apt/lists/* # buildkit
2026-05-08 20:14:43 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
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)ENV UV_HTTP_TIMEOUT=500
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)ENV UV_INDEX_STRATEGY=unsafe-best-match
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)ENV UV_LINK_MODE=copy
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)ENV VLLM_ENABLE_CUDA_COMPATIBILITY=0
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)ARG PYTORCH_CUDA_INDEX_BASE_URL
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)COPY requirements/common.txt /tmp/common.txt # buildkit
2026-05-08 20:14:43 UTC (buildkit.dockerfile.v0)COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit
2026-05-08 20:15:07 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 $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && rm /tmp/requirements-cuda.txt /tmp/common.txt # buildkit
2026-05-08 20:15:07 UTC (buildkit.dockerfile.v0)ARG FLASHINFER_VERSION=0.6.8.post1
2026-05-08 20:22:14 UTC (buildkit.dockerfile.v0)RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl FLASHINFER_VERSION=0.6.8.post1 /bin/sh -c uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} --extra-index-url https://flashinfer.ai/whl/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && flashinfer show-config && flashinfer download-cubin # buildkit
2026-05-08 20:22:14 UTC (buildkit.dockerfile.v0)ARG GDRCOPY_CUDA_VERSION=12.8
2026-05-08 20:22:14 UTC (buildkit.dockerfile.v0)ARG GDRCOPY_OS_VERSION=Ubuntu22_04
2026-05-08 20:22:14 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM
2026-05-08 20:22:14 UTC (buildkit.dockerfile.v0)COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit
2026-05-08 20:22:19 UTC (buildkit.dockerfile.v0)RUN |10 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl FLASHINFER_VERSION=0.6.8.post1 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/amd64 /bin/sh -c set -eux; case "${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
2026-05-08 20:22:19 UTC (buildkit.dockerfile.v0)ARG BITSANDBYTES_VERSION_X86=0.46.1
2026-05-08 20:22:19 UTC (buildkit.dockerfile.v0)ARG BITSANDBYTES_VERSION_ARM64=0.42.0
2026-05-08 20:22:19 UTC (buildkit.dockerfile.v0)ARG TIMM_VERSION=>=1.0.17
2026-05-08 20:22:19 UTC (buildkit.dockerfile.v0)ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.7
2026-05-08 20:22:23 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=0.6.8.post1 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
2026-05-08 20:22:23 UTC (buildkit.dockerfile.v0)ARG PIP_INDEX_URL UV_INDEX_URL
2026-05-08 20:22:23 UTC (buildkit.dockerfile.v0)ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL
2026-05-08 20:22:23 UTC (buildkit.dockerfile.v0)ARG PYTORCH_CUDA_INDEX_BASE_URL
2026-05-08 20:22:23 UTC (buildkit.dockerfile.v0)ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER
2026-05-08 20:22:23 UTC (buildkit.dockerfile.v0)ARG PYTORCH_NIGHTLY
2026-05-08 20:22:23 UTC (buildkit.dockerfile.v0)COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit
2026-05-08 20:29:37 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=0.6.8.post1 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
2026-05-08 20:29:38 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=0.6.8.post1 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
2026-05-08 20:29:38 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64
2026-05-08 20:29:39 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=0.6.8.post1 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 $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') # buildkit
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)COPY examples examples # buildkit
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)COPY benchmarks benchmarks # buildkit
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)COPY ./vllm/collect_env.py . # buildkit
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ARG INSTALL_KV_CONNECTORS=false
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ARG PIP_INDEX_URL UV_INDEX_URL
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ENV UV_HTTP_TIMEOUT=500
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ARG torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 11.0 12.0+PTX
2026-05-08 20:29:39 UTC (buildkit.dockerfile.v0)ENV TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 8.9 9.0 10.0 12.0
2026-05-08 20:29:46 UTC (buildkit.dockerfile.v0)RUN |8 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=12.9.1 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 12.0 /bin/sh -c CUDA_MAJOR="${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 --allow-change-held-packages ${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
2026-05-08 20:29:46 UTC (buildkit.dockerfile.v0)ENV VLLM_USAGE_SOURCE=production-docker-image
2026-05-08 20:29:46 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["vllm" "serve"]
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/amd64
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=amd64
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ENV DEBIAN_FRONTEND=noninteractive LANG=en_US.UTF-8 LANGUAGE=en_US:en LC_ALL=en_US.UTF-8
2026-05-25 12:20:22 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
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ARG PYTHON_VERSION=3.12
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ENV PYTHON_VERSION=3.12
2026-05-25 12:20:22 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 ! command -v python >/dev/null 2>&1 \ && command -v python${PYTHON_VERSION} >/dev/null 2>&1; then update-alternatives --install /usr/bin/python python $(which python${PYTHON_VERSION}) 1 fi # buildkit
2026-05-25 12:20:22 UTC (buildkit.dockerfile.v0)ENV PIP_NO_CACHE_DIR=1 PIP_DISABLE_PIP_VERSION_CHECK=1 PIP_ROOT_USER_ACTION=ignore
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)RUN |4 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 /bin/bash -eo pipefail -c # Buildkit cat <<EOT >/tmp/requirements.txt build cmake<4 wheel pybind11<3 Cython EOT pip install -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION=12.9.1
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG CUDA_ARCHS
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ENV CUDA_HOME=/usr/local/cuda CUDA_VERSION=12.9.1 CUDA_ARCHS=
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ENV UV_SYSTEM_PYTHON=1 UV_PRERELEASE=allow UV_SKIP_WHEEL_FILENAME_CHECK=1 UV_NO_CACHE=1 UV_HTTP_TIMEOUT=500 UV_INDEX_STRATEGY=unsafe-best-match
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG VLLM_VERSION=0.20.2
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG VLLM_TORCH_VERSION=2.11.0
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG VLLM_TORCH_CUDA_VERSION=12.9.1
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ENV VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 VLLM_TORCH_CUDA_VERSION=12.9.1
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/amd64
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-05-25 12:20:25 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=amd64
2026-05-25 12:20:28 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Ray # Install Ray Client and Default RAY_VERSION=2.55.1 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
2026-05-25 12:21:13 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 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
2026-05-25 12:21:13 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 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
2026-05-25 12:21:15 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 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 $(pip show vllm_omni | grep Location: | cut -d" " -f 2) \ && patch -p1 < /workspace/patches/vllm_omni/*.patch; \ fi # buildkit
2026-05-25 12:21:16 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Dependencies # Install cat <<EOT >/tmp/requirements.txt # audio extras librosa soundfile mistral_common[audio] EOT uv pip install \ -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-05-25 12:21:16 UTC (buildkit.dockerfile.v0)ENV RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES=1 SAFETENSORS_FAST_GPU=1 RUNAI_STREAMER_MEMORY_LIMIT=0 RUNAI_STREAMER_LOG_TO_STDERR=1 RUNAI_STREAMER_LOG_LEVEL=INFO
2026-05-25 12:21:16 UTC (buildkit.dockerfile.v0)WORKDIR /
2026-05-25 12:21:16 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["tini" "--"]
2026-06-05 14:06:31 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-06-05 14:06:31 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/amd64
2026-06-05 14:06:31 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-06-05 14:06:31 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=amd64
2026-06-05 14:06:31 UTC (buildkit.dockerfile.v0)RUN |3 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 /bin/bash -eo pipefail -c # Patch F="$(pip show vllm | grep Location: | cut -d" " -f 2)/vllm/utils/network_utils.py" echo "Target: ${F}" grep -n "_next_port" "${F}" || true if grep -qE '^[[:space:]]*_next_port = port \+ 1[[:space:]]*$' "${F}"; then echo "Already fixed (_next_port = port + 1), skipping." elif grep -qE '^[[:space:]]*_next_port = port[[:space:]]*$' "${F}"; then sed -i -E 's/^([[:space:]]*)_next_port = port[[:space:]]*$/\1_next_port = port + 1/' "${F}" echo "Patched: _next_port = port -> _next_port = port + 1" else echo "ERROR: expected '_next_port = port' assignment not found in ${F}" >&2 exit 1 fi grep -n "_next_port" "${F}" # buildkit
2026-06-05 14:06:40 UTC (buildkit.dockerfile.v0)RUN |3 TARGETPLATFORM=linux/amd64 TARGETOS=linux TARGETARCH=amd64 /bin/bash -eo pipefail -c # Pin Ray pip install --no-cache-dir ray==2.54.0 pip show ray | grep -i '^Version:' || true # buildkit
2026-06-05 14:06:40 UTC (buildkit.dockerfile.v0)WORKDIR /
2026-06-05 14:06:40 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["tini" "--"]
2026-06-05 14:07:14 UTC
11.4 GB
/usr/local/cuda
CUDA_VERSION12.9.1
DEBIAN_FRONTENDnoninteractive
LANGen_US.UTF-8
LANGUAGEen_US:en
LC_ALLen_US.UTF-8
LD_LIBRARY_PATH/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64
NVARCHsbsa
NVIDIA_DRIVER_CAPABILITIEScompute,utility
NVIDIA_REQUIRE_CUDAcuda>=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_DEVICESall
NV_CUDA_CUDART_VERSION12.9.79-1
PATH/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
PIP_DISABLE_PIP_VERSION_CHECK1
PIP_NO_CACHE_DIR1
PIP_ROOT_USER_ACTIONignore
PYTHON_VERSION3.12
RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES1
RUNAI_STREAMER_LOG_LEVELINFO
RUNAI_STREAMER_LOG_TO_STDERR1
RUNAI_STREAMER_MEMORY_LIMIT0
SAFETENSORS_FAST_GPU1
TORCH_CUDA_ARCH_LIST8.0 8.7 8.9 9.0 10.0 12.0
UV_HTTP_TIMEOUT500
UV_INDEX_STRATEGYunsafe-best-match
UV_LINK_MODEcopy
UV_NO_CACHE1
UV_PRERELEASEallow
UV_SKIP_WHEEL_FILENAME_CHECK1
UV_SYSTEM_PYTHON1
VLLM_ENABLE_CUDA_COMPATIBILITY0
VLLM_TORCH_CUDA_VERSION12.9.1
VLLM_TORCH_VERSION2.11.0
VLLM_USAGE_SOURCEproduction-docker-image
VLLM_VERSION0.20.2
[#000] sha256:1fe4490d3380a3fec7f0b5cfd536dbdeed91ff50c5fbd11eba4ec838d689855e - 0.25% (28.6 MB)
[#001] sha256:ecbca12713d0ee06710d449b9cecb86e2b3be1c32e2f7e6fd6ff3dea9219cf03 - 0.04% (4.32 MB)
[#002] sha256:ed5ca6b9e891c487cf1c2321e0b3e28f63b7ffdc2ee7c75592578bfcdebac740 - 0.8% (93.1 MB)
[#003] sha256:3f0047ea99ac7fb8902bd5e3aeb361d80ffa04c40b26b7d3935b6709c3c66f0c - 0.0% (182 Bytes)
[#004] sha256:6e8af4fd0a071982e528b634ba99dec2474c21147f99748be708f36e10e3f4c2 - 0.0% (6.72 KB)
[#005] sha256:8c332ece29a4ef3af19ad1cfb84e1f85bde014f649b71ffd984a78a7e55f0457 - 0.0% (104 Bytes)
[#006] sha256:549e915cec8d41c0d000dc3fd4ea0fad54182ae2011d6a3e2af4e8e8ca3a7bb6 - 0.0% (230 Bytes)
[#007] sha256:a9ba02214da23c6a06716b6f5212e0ce90df7c54bae34bf2d574dfb70e0dda4f - 3.24% (377 MB)
[#008] sha256:8e5da31cf2afdd88f72605e3b4a82f9da757eed24ea0c2d62d23371ccde9b521 - 12.23% (1.39 GB)
[#009] sha256:2bc8a2f5e65083f212b6c680d1c39862bf89480383dbf0bbff961e637f2c3e3a - 0.39% (45.3 MB)
[#010] sha256:989f48b0519e1f859723b0ac8e569a5076bfa3ae208bed5322f5a8c54ca3a67d - 0.0% (1.52 KB)
[#011] sha256:48da5a62e9c562c81050d972acfbd65463f69ddf4112f9ea4c7f7d517507ad4f - 0.0% (628 Bytes)
[#012] sha256:6da5c45484246d5c04e2d0b162f0e59061e2ddf76b9fdf2062cbe2bb3a32ed98 - 39.28% (4.47 GB)
[#013] sha256:49eba36e19194e7ac0626e446c11260bdc9aebbea2edbcaa2ea90a7140dd141a - 18.04% (2.05 GB)
[#014] sha256:329a5bcd66f5d0f4a550b52f6563f8deee698b278926be2c77c9b461b8c7856b - 0.0% (894 Bytes)
[#015] sha256:30e10778d4b0c3e9c6187e05eb20d00f0ccde68eca13a3b53e6a8e566f275fde - 0.0% (487 KB)
[#016] sha256:24e83a38862f528e432ddb55ca1c02a18ad0bd584158cc236cbd349db4ae27e4 - 3.1% (360 MB)
[#017] sha256:e8500d2e2d5f60dc2ee68a4daeeb610e7beb912c952a5053069e0f825042c346 - 0.0% (200 Bytes)
[#018] sha256:110e0b3116a050b7ff5d8522bdc9e6986f7332146cab757af74c88f01a2b8597 - 5.77% (672 MB)
[#019] sha256:53fd89ecdac86cd7b64234843a0f0c1410ce3c8244efb53fdd6da0be181b8a95 - 0.0% (123 Bytes)
[#020] sha256:21619eac2d83ee1326c1aee418d8c66b3776e2ac69591c77aaf6e6991f3c2aed - 0.13% (15.2 MB)
[#021] sha256:08a03d1ed3818403dad6250968d59ce2cd912cdd0641001efff4b28a3cf5eaf5 - 0.0% (241 KB)
[#022] sha256:103895fdc9ae488a21f885881a1fcc8c9e0c4a62f65ca43e106538d0d5eee66e - 0.0% (224 KB)
[#023] sha256:9e2b3cbd219d4fcdab83bc89554532e061ddd34e6aae25283e1034bad9c8105a - 0.0% (9.62 KB)
[#024] sha256:08886895dc0ef81ff50562a887caa34da75065981c31ab0baf5099589f310e96 - 10.58% (1.2 GB)
[#025] sha256:d273de6675902c86b66e5777e47c1a2e7bd05ea8b77794f567615a52b08242be - 1.09% (127 MB)
[#026] sha256:81f85958c7d9512115cfbe391aec6b821e487fa5bb67ef9ca6226787076dd1f8 - 0.0% (1.87 KB)
[#027] sha256:c4863656d03a186260967d48df8be43ec90183be1e4bcfc45a4d6c02c9ed055a - 0.26% (30.8 MB)
[#028] sha256:ffbd971c766cce7659aeec4065b15f924e11a73be656ca3edcc6c28153a7aaf9 - 0.69% (80.8 MB)
[#029] sha256:644a9af89afd4c73bf5c65aefa7355a19f737fb37cafef8b6c13ae4137cc8e00 - 3.35% (390 MB)
[#030] sha256:4f4fb700ef54461cfa02571ae0db9a0dc1e0cdb5577484a6d75e68dc38e8acc1 - 0.0% (32 Bytes)
[#031] sha256:684436c3eef08af37c61cdc33a81b08c12c466e4a29c93acd7048202da71d5fc - 0.0% (6.19 KB)
[#032] sha256:9be858d9423791552f175e83b7a8d160b1b0120d5a2c4e1f692ec948979d583b - 0.08% (9.33 MB)
[#033] sha256:f1502e7d513b4d30f9ca8a7388f7456838cc77969b762e0034f9e87487453fb0 - 0.0% (4.04 KB)
[#034] sha256:7acdf8ea6735c0460afa55e84c1f02b20ed22fbc947faf41a778aa1ddb2ad4cd - 0.67% (78.4 MB)
/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=${NV_CUDA_CUDART_VERSION} cuda-compat-12-9 && rm -rf /var/lib/apt/lists/* # buildkit
2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)RUN |1 TARGETARCH=arm64 /bin/sh -c echo "/usr/local/cuda/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit
2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64
2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)COPY NGC-DL-CONTAINER-LICENSE / # buildkit
2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)ENV NVIDIA_VISIBLE_DEVICES=all
2025-07-18 20:20:54 UTC (buildkit.dockerfile.v0)ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)ARG PYTHON_VERSION
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)ARG DEADSNAKES_MIRROR_URL
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)ARG DEADSNAKES_GPGKEY_URL
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)ARG GET_PIP_URL
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)ENV DEBIAN_FRONTEND=noninteractive
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)WORKDIR /vllm-workspace
2026-05-08 20:09:54 UTC (buildkit.dockerfile.v0)RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c PYTHON_VERSION_STR=$(echo ${PYTHON_VERSION} | sed 's/\.//g') && echo "export PYTHON_VERSION_STR=${PYTHON_VERSION_STR}" >> /etc/environment # buildkit
2026-05-08 20:13:27 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 apt-get update -y && apt-get install -y --no-install-recommends curl sudo ffmpeg libsm6 libxext6 libgl1 libibverbs-dev build-essential libssl-dev libffi-dev zlib1g-dev libbz2-dev libreadline-dev libsqlite3-dev libncurses-dev liblzma-dev libgdbm-dev uuid-dev tk-dev && PYTHON_MAJOR_MINOR=${PYTHON_VERSION} && PYTHON_FULL_VERSION=$(curl -s https://www.python.org/ftp/python/ | grep -oE "${PYTHON_MAJOR_MINOR}\.[0-9]+" | sort -t. -k3 -n | uniq | tail -1) && echo "Building Python ${PYTHON_FULL_VERSION} from source..." && curl -fSL https://www.python.org/ftp/python/${PYTHON_FULL_VERSION}/Python-${PYTHON_FULL_VERSION}.tgz -o /tmp/python.tgz && tar -xzf /tmp/python.tgz -C /tmp && cd /tmp/Python-${PYTHON_FULL_VERSION} && ./configure --enable-optimizations --with-ensurepip=install --prefix=/usr/local && make -j$(nproc) && make install && cd / && rm -rf /tmp/python.tgz /tmp/Python-${PYTHON_FULL_VERSION} && ln -sf /usr/local/bin/python${PYTHON_MAJOR_MINOR} /usr/bin/python3 && ln -sf /usr/local/bin/python${PYTHON_MAJOR_MINOR}-config /usr/bin/python3-config && ln -sf /usr/local/bin/pip${PYTHON_MAJOR_MINOR} /usr/bin/pip && rm -rf /var/lib/apt/lists/* && python3 --version && python3 -m pip --version # buildkit
2026-05-08 20:14:16 UTC (buildkit.dockerfile.v0)RUN |5 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py /bin/sh -c CUDA_VERSION_DASH=$(echo $CUDA_VERSION | cut -d. -f1,2 | tr '.' '-') && CUDA_VERSION_SHORT=$(echo $CUDA_VERSION | cut -d. -f1,2) && apt-get update -y && apt-get install -y --no-install-recommends --allow-change-held-packages cuda-nvcc-${CUDA_VERSION_DASH} cuda-cudart-${CUDA_VERSION_DASH} cuda-nvrtc-${CUDA_VERSION_DASH} cuda-cuobjdump-${CUDA_VERSION_DASH} libcurand-dev-${CUDA_VERSION_DASH} libcublas-${CUDA_VERSION_DASH} libnuma-dev numactl && NCCL_VER=$(apt-cache madison libnccl-dev | grep "+cuda${CUDA_VERSION_SHORT}" | head -1 | awk -F'|' '{gsub(/^ +| +$/, "", $2); print $2}') && apt-get install -y --no-install-recommends --allow-change-held-packages libnccl-dev=${NCCL_VER} libnccl2=${NCCL_VER} && rm -rf /var/lib/apt/lists/* # buildkit
2026-05-08 20:14:18 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
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)ENV UV_HTTP_TIMEOUT=500
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)ENV UV_INDEX_STRATEGY=unsafe-best-match
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)ENV UV_LINK_MODE=copy
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)ENV VLLM_ENABLE_CUDA_COMPATIBILITY=0
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)ARG PYTORCH_CUDA_INDEX_BASE_URL
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)COPY requirements/common.txt /tmp/common.txt # buildkit
2026-05-08 20:14:18 UTC (buildkit.dockerfile.v0)COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit
2026-05-08 20:14:30 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 $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && rm /tmp/requirements-cuda.txt /tmp/common.txt # buildkit
2026-05-08 20:14:30 UTC (buildkit.dockerfile.v0)ARG FLASHINFER_VERSION=0.6.8.post1
2026-05-08 20:21:20 UTC (buildkit.dockerfile.v0)RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl FLASHINFER_VERSION=0.6.8.post1 /bin/sh -c uv pip install --system flashinfer-jit-cache==${FLASHINFER_VERSION} --extra-index-url https://flashinfer.ai/whl/cu$(echo $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') && flashinfer show-config && flashinfer download-cubin # buildkit
2026-05-08 20:21:20 UTC (buildkit.dockerfile.v0)ARG GDRCOPY_CUDA_VERSION=12.8
2026-05-08 20:21:20 UTC (buildkit.dockerfile.v0)ARG GDRCOPY_OS_VERSION=Ubuntu22_04
2026-05-08 20:21:20 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM
2026-05-08 20:21:20 UTC (buildkit.dockerfile.v0)COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit
2026-05-08 20:21:34 UTC (buildkit.dockerfile.v0)RUN |10 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL= DEADSNAKES_GPGKEY_URL= GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl FLASHINFER_VERSION=0.6.8.post1 GDRCOPY_CUDA_VERSION=12.8 GDRCOPY_OS_VERSION=Ubuntu24_04 TARGETPLATFORM=linux/arm64 /bin/sh -c set -eux; case "${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
2026-05-08 20:21:34 UTC (buildkit.dockerfile.v0)ARG BITSANDBYTES_VERSION_X86=0.46.1
2026-05-08 20:21:34 UTC (buildkit.dockerfile.v0)ARG BITSANDBYTES_VERSION_ARM64=0.42.0
2026-05-08 20:21:34 UTC (buildkit.dockerfile.v0)ARG TIMM_VERSION=>=1.0.17
2026-05-08 20:21:34 UTC (buildkit.dockerfile.v0)ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.7
2026-05-08 20:21:38 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=0.6.8.post1 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
2026-05-08 20:21:38 UTC (buildkit.dockerfile.v0)ARG PIP_INDEX_URL UV_INDEX_URL
2026-05-08 20:21:38 UTC (buildkit.dockerfile.v0)ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL
2026-05-08 20:21:38 UTC (buildkit.dockerfile.v0)ARG PYTORCH_CUDA_INDEX_BASE_URL
2026-05-08 20:21:38 UTC (buildkit.dockerfile.v0)ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER
2026-05-08 20:21:38 UTC (buildkit.dockerfile.v0)ARG PYTORCH_NIGHTLY
2026-05-08 20:21:38 UTC (buildkit.dockerfile.v0)COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit
2026-05-08 20:28:37 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=0.6.8.post1 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
2026-05-08 20:28:37 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=0.6.8.post1 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
2026-05-08 20:28:37 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64
2026-05-08 20:28:38 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=0.6.8.post1 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 $CUDA_VERSION | cut -d. -f1,2 | tr -d '.') # buildkit
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)COPY examples examples # buildkit
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)COPY benchmarks benchmarks # buildkit
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)COPY ./vllm/collect_env.py . # buildkit
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ARG INSTALL_KV_CONNECTORS=false
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ARG PIP_INDEX_URL UV_INDEX_URL
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ENV UV_HTTP_TIMEOUT=500
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ARG torch_cuda_arch_list=7.5 8.0 8.6 8.9 9.0 10.0 11.0 12.0+PTX
2026-05-08 20:28:38 UTC (buildkit.dockerfile.v0)ENV TORCH_CUDA_ARCH_LIST=8.0 8.7 8.9 9.0 10.0 12.0
2026-05-08 20:35:44 UTC (buildkit.dockerfile.v0)RUN |8 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=12.9.1 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL= UV_EXTRA_INDEX_URL= torch_cuda_arch_list=8.0 8.7 8.9 9.0 10.0 12.0 /bin/sh -c CUDA_MAJOR="${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 --allow-change-held-packages ${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
2026-05-08 20:35:44 UTC (buildkit.dockerfile.v0)ENV VLLM_USAGE_SOURCE=production-docker-image
2026-05-08 20:35:44 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["vllm" "serve"]
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/arm64
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=arm64
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ENV DEBIAN_FRONTEND=noninteractive LANG=en_US.UTF-8 LANGUAGE=en_US:en LC_ALL=en_US.UTF-8
2026-05-25 12:20:44 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
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ARG PYTHON_VERSION=3.12
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ENV PYTHON_VERSION=3.12
2026-05-25 12:20:44 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 ! command -v python >/dev/null 2>&1 \ && command -v python${PYTHON_VERSION} >/dev/null 2>&1; then update-alternatives --install /usr/bin/python python $(which python${PYTHON_VERSION}) 1 fi # buildkit
2026-05-25 12:20:44 UTC (buildkit.dockerfile.v0)ENV PIP_NO_CACHE_DIR=1 PIP_DISABLE_PIP_VERSION_CHECK=1 PIP_ROOT_USER_ACTION=ignore
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)RUN |4 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 /bin/bash -eo pipefail -c # Buildkit cat <<EOT >/tmp/requirements.txt build cmake<4 wheel pybind11<3 Cython EOT pip install -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG CUDA_VERSION=12.9.1
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG CUDA_ARCHS
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ENV CUDA_HOME=/usr/local/cuda CUDA_VERSION=12.9.1 CUDA_ARCHS=
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ENV UV_SYSTEM_PYTHON=1 UV_PRERELEASE=allow UV_SKIP_WHEEL_FILENAME_CHECK=1 UV_NO_CACHE=1 UV_HTTP_TIMEOUT=500 UV_INDEX_STRATEGY=unsafe-best-match
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG VLLM_VERSION=0.20.2
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG VLLM_TORCH_VERSION=2.11.0
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG VLLM_TORCH_CUDA_VERSION=12.9.1
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ENV VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 VLLM_TORCH_CUDA_VERSION=12.9.1
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/arm64
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-05-25 12:20:47 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=arm64
2026-05-25 12:20:49 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Ray # Install Ray Client and Default RAY_VERSION=2.55.1 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
2026-05-25 12:21:29 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 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
2026-05-25 12:21:29 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 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
2026-05-25 12:21:31 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 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 $(pip show vllm_omni | grep Location: | cut -d" " -f 2) \ && patch -p1 < /workspace/patches/vllm_omni/*.patch; \ fi # buildkit
2026-05-25 12:21:32 UTC (buildkit.dockerfile.v0)RUN |9 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 PYTHON_VERSION=3.12 CUDA_VERSION=12.9.1 CUDA_ARCHS= VLLM_VERSION=0.20.2 VLLM_TORCH_VERSION=2.11.0 VLLM_TORCH_CUDA_VERSION=12.9.1 /bin/bash -eo pipefail -c # Dependencies # Install cat <<EOT >/tmp/requirements.txt # audio extras librosa soundfile mistral_common[audio] EOT uv pip install \ -r /tmp/requirements.txt # Cleanup rm -rf /var/tmp/* \ && rm -rf /tmp/* # buildkit
2026-05-25 12:21:32 UTC (buildkit.dockerfile.v0)ENV RAY_EXPERIMENTAL_NOSET_CUDA_VISIBLE_DEVICES=1 SAFETENSORS_FAST_GPU=1 RUNAI_STREAMER_MEMORY_LIMIT=0 RUNAI_STREAMER_LOG_TO_STDERR=1 RUNAI_STREAMER_LOG_LEVEL=INFO
2026-05-25 12:21:32 UTC (buildkit.dockerfile.v0)WORKDIR /
2026-05-25 12:21:32 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["tini" "--"]
2026-06-05 14:07:07 UTC (buildkit.dockerfile.v0)SHELL [/bin/bash -eo pipefail -c]
2026-06-05 14:07:07 UTC (buildkit.dockerfile.v0)ARG TARGETPLATFORM=linux/arm64
2026-06-05 14:07:07 UTC (buildkit.dockerfile.v0)ARG TARGETOS=linux
2026-06-05 14:07:07 UTC (buildkit.dockerfile.v0)ARG TARGETARCH=arm64
2026-06-05 14:07:07 UTC (buildkit.dockerfile.v0)RUN |3 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 /bin/bash -eo pipefail -c # Patch F="$(pip show vllm | grep Location: | cut -d" " -f 2)/vllm/utils/network_utils.py" echo "Target: ${F}" grep -n "_next_port" "${F}" || true if grep -qE '^[[:space:]]*_next_port = port \+ 1[[:space:]]*$' "${F}"; then echo "Already fixed (_next_port = port + 1), skipping." elif grep -qE '^[[:space:]]*_next_port = port[[:space:]]*$' "${F}"; then sed -i -E 's/^([[:space:]]*)_next_port = port[[:space:]]*$/\1_next_port = port + 1/' "${F}" echo "Patched: _next_port = port -> _next_port = port + 1" else echo "ERROR: expected '_next_port = port' assignment not found in ${F}" >&2 exit 1 fi grep -n "_next_port" "${F}" # buildkit
2026-06-05 14:07:14 UTC (buildkit.dockerfile.v0)RUN |3 TARGETPLATFORM=linux/arm64 TARGETOS=linux TARGETARCH=arm64 /bin/bash -eo pipefail -c # Pin Ray pip install --no-cache-dir ray==2.54.0 pip show ray | grep -i '^Version:' || true # buildkit
2026-06-05 14:07:14 UTC (buildkit.dockerfile.v0)WORKDIR /
2026-06-05 14:07:14 UTC (buildkit.dockerfile.v0)ENTRYPOINT ["tini" "--"]
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.