Sign inSign up

velaraptor1/vllm-benchpod:shim-qwen38-x86_64-cu129

Index digest

sha256:8d328a7e4b570ae978cafa88871718c0c7ccc586d23949a2d4557b9638e9dfb6

OS/ARCH

linux/amd64

Compressed size

9.2 GB

Last pushed

about 1 month by velaraptor1

Type

Image

Manifest digest

sha256:b02fef0c5e75b0da580bf77945873946ea3769bfa8b8688dfe861fd10aa0ba30

Image Layers

Image Layers

1ARG RELEASE0 B
2ARG LAUNCHPAD_BUILD_ARCH0 B
3LABEL org.opencontainers.image.ref.name=ubuntu0 B
4LABEL org.opencontainers.image.version=24.040 B
5ADD file ... in / 29.17 MB
6CMD ["/bin/bash"]0 B
7ENV NVARCH=x86_640 B
8ENV NVIDIA_REQUIRE_CUDA=cuda>=12.9 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<5360 B
9ENV NV_CUDA_CUDART_VERSION=12.9.79-10 B
10ARG TARGETARCH0 B
11LABEL maintainer=NVIDIA CORPORATION <[email protected]>0 B
12RUN |1 TARGETARCH=amd64 /bin/sh -c4.34 MB
13ENV CUDA_VERSION=12.9.10 B
14RUN |1 TARGETARCH=amd64 /bin/sh -c98.72 MB
15RUN |1 TARGETARCH=amd64 /bin/sh -c182 B
16ENV PATH=/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B
17ENV LD_LIBRARY_PATH=/usr/local/cuda/lib640 B
18COPY NGC-DL-CONTAINER-LICENSE / # buildkit6.72 KB
19ENV NVIDIA_VISIBLE_DEVICES=all0 B
20ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility0 B
21ARG CUDA_VERSION=12.9.10 B
22ARG PYTHON_VERSION=3.120 B
23ARG NCCL_VERSION=2.30.70 B
24ARG DEADSNAKES_MIRROR_URL0 B
25ARG DEADSNAKES_GPGKEY_URL0 B
26ARG GET_PIP_URL=https://bootstrap.pypa.io/get-pip.py0 B
27ENV DEBIAN_FRONTEND=noninteractive0 B
28WORKDIR /vllm-workspace102 B
29RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7230 B
30RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7222.55 MB
31RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.71.58 GB
32RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.743.15 MB
33ENV UV_HTTP_TIMEOUT=5000 B
34ENV UV_INDEX_STRATEGY=unsafe-best-match0 B
35ENV UV_LINK_MODE=copy0 B
36ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python0 B
37ENV UV_CACHE_DIR=/opt/uv/cache0 B
38RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7150 B
39RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7163 B
40ENV UV_OVERRIDE=/etc/uv-overrides.txt0 B
41RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.73.4 KB
42COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB
43RUN |6 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.732 B
44ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B
45ARG PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl0 B
46COPY requirements/common.txt /tmp/common.txt # buildkit1.63 KB
47COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit776 B
48RUN |7 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.74.55 GB
49ARG FLASHINFER_VERSION=0.6.170 B
50RUN |8 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.71.81 GB
51COPY docker/patches/flashinfer-0.6.17-pr-4266.patch docker/patches/flashinfer-0.6.17-pr-4455.patch /tmp/ #14.51 KB
52RUN |8 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.769.76 KB
53ARG GDRCOPY_CUDA_VERSION=12.80 B
54ARG GDRCOPY_OS_VERSION=Ubuntu24_040 B
55ARG TARGETPLATFORM=linux/amd640 B
56COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit897 B
57RUN |11 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7320.4 KB
58ARG TIMM_VERSION=>=1.0.170 B
59ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
60RUN |13 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.758.6 MB
61ARG PIP_INDEX_URL UV_INDEX_URL=0 B
62ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL=0 B
63ARG PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl0 B
64ARG PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled0 B
65ARG PYTORCH_NIGHTLY0 B
66COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit203 B
67COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit237 B
68RUN |20 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7516.17 MB
69RUN |20 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.732 B
70ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib640 B
71COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit254 B
72RUN |20 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.718.19 MB
73ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib640 B
74COPY examples examples # buildkit264.59 KB
75COPY benchmarks benchmarks # buildkit275.24 KB
76COPY ./vllm/collect_env.py . # buildkit9.61 KB
77ARG TARGETPLATFORM=linux/amd640 B
78ARG INSTALL_KV_CONNECTORS=true0 B
79ARG CUDA_VERSION=12.9.10 B
80ARG VLLM_BUILD_COMMIT=3a0914114705fa38d4c3171d0746c1a6b6f102090 B
81ARG VLLM_BUILD_PIPELINE0 B
82ARG VLLM_BUILD_URL0 B
83ARG VLLM_IMAGE_TAG=inferactinc/dev:wtn-3a09141-x86_64-cu12.9.10 B
84ARG PIP_INDEX_URL UV_INDEX_URL=0 B
85ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL=0 B
86ENV UV_HTTP_TIMEOUT=5000 B
87ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
88ENV TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 8.90 B
89RUN |26 CUDA_VERSION=12.9.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7295.02 MB
90ENV VLLM_USAGE_SOURCE=production-docker-image0 B
91ENV VLLM_BUILD_COMMIT=3a0914114705fa38d4c3171d0746c1a6b6f10209 VLLM_BUILD_PIPELINE=local VLLM_BUILD_URL= VLLM_IMAGE_TAG=inferactinc/dev:wtn-3a09141-x86_64-cu12.9.10 B
92LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=3a0914114705fa38d4c3171d0746c1a6b6f10209 org.opencontainers.image.version=inferactinc/dev:wtn-3a09141-x86_64-cu12.9.1 org.opencontainers.image.url=0 B
93ENTRYPOINT ["vllm" "serve"]0 B
94COPY bootstrap.sh /bootstrap.sh # buildkit2.66 KB
95ENTRYPOINT []0 B
96CMD ["bash" "/bootstrap.sh"]0 B

Command

 ARG RELEASE