Sign inSign up

pt9912/vllm-qwen38-b200:1

Machine learning & AI
Manifest digest

sha256:6f13de2fb100914c5ea5d381bc8cdb98a9b9ceeb9df7d9435197e7fb92d974dd

OS/ARCH

linux/amd64

Compressed size

8.07 GB

Last pushed

23 days by pt9912

Type

Image

Manifest digest

sha256:6f13de2fb100914c5ea5d381bc8cdb98a9b9ceeb9df7d9435197e7fb92d974dd

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 / 29.18 MB
6CMD ["/bin/bash"]0 B
7ENV NVARCH=x86_640 B
8ENV NVIDIA_REQUIRE_CUDA=cuda>=13.0 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<5360 B
9ENV NV_CUDA_CUDART_VERSION=13.0.88-10 B
10ARG TARGETARCH0 B
11LABEL maintainer=NVIDIA CORPORATION <[email protected]>0 B
12RUN |1 TARGETARCH=amd64 /bin/sh -c4.34 MB
13ENV CUDA_VERSION=13.0.10 B
14RUN |1 TARGETARCH=amd64 /bin/sh -c100.28 MB
15RUN |1 TARGETARCH=amd64 /bin/sh -c184 B
16ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B
17ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/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=13.0.10 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-workspace104 B
29RUN |6 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7228 B
30RUN |6 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7222.46 MB
31RUN |6 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.71.28 GB
32RUN |6 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.740.49 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=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7150 B
39RUN |6 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7163 B
40ENV UV_OVERRIDE=/etc/uv-overrides.txt0 B
41RUN |6 CUDA_VERSION=13.0.1 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=13.0.1 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 # buildkit865 B
48RUN |7 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.74.32 GB
49ARG FLASHINFER_VERSION=0.6.170 B
50RUN |8 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.71.41 GB
51ARG GDRCOPY_CUDA_VERSION=12.80 B
52ARG GDRCOPY_OS_VERSION=Ubuntu24_040 B
53ARG TARGETPLATFORM=linux/amd640 B
54COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit897 B
55RUN |11 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7321.43 KB
56ARG TIMM_VERSION=>=1.0.170 B
57ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
58RUN |13 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.758.72 MB
59ARG PIP_INDEX_URL UV_INDEX_URL=0 B
60ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL=0 B
61ARG PYTORCH_CUDA_INDEX_BASE_URL=https://download.pytorch.org/whl0 B
62ARG PIP_KEYRING_PROVIDER=disabled UV_KEYRING_PROVIDER=disabled0 B
63ARG PYTORCH_NIGHTLY0 B
64COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit201 B
65COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit238 B
66RUN |20 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7295.34 MB
67RUN |20 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.732 B
68ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
69COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit254 B
70RUN |20 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.713.81 MB
71ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
72COPY examples examples # buildkit263.4 KB
73COPY benchmarks benchmarks # buildkit283.25 KB
74COPY ./vllm/collect_env.py . # buildkit9.61 KB
75ARG TARGETPLATFORM=linux/amd640 B
76ARG INSTALL_KV_CONNECTORS=true0 B
77ARG CUDA_VERSION=13.0.10 B
78ARG VLLM_BUILD_COMMIT0 B
79ARG VLLM_BUILD_PIPELINE0 B
80ARG VLLM_BUILD_URL0 B
81ARG VLLM_IMAGE_TAG0 B
82ARG PIP_INDEX_URL UV_INDEX_URL=0 B
83ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL=0 B
84ENV UV_HTTP_TIMEOUT=5000 B
85ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
86ENV TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 8.90 B
87RUN |26 CUDA_VERSION=13.0.1 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7296.43 MB
88ENV VLLM_USAGE_SOURCE=production-docker-image0 B
89ENV VLLM_BUILD_COMMIT=unknown VLLM_BUILD_PIPELINE=local VLLM_BUILD_URL= VLLM_IMAGE_TAG=local/vllm-openai:dev0 B
90LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=unknown org.opencontainers.image.version=local/vllm-openai:dev org.opencontainers.image.url=0 B
91ENTRYPOINT ["vllm" "serve"]0 B
92USER root0 B
93ARG RECIPE_REF=17c58984378af20fcec1f4ef3e96f244493d48e50 B
94RUN |1 RECIPE_REF=17c58984378af20fcec1f4ef3e96f244493d48e5 /bin/sh -c9.04 MB
95RUN |1 RECIPE_REF=17c58984378af20fcec1f4ef3e96f244493d48e5 /bin/sh -c114.69 KB
96RUN |1 RECIPE_REF=17c58984378af20fcec1f4ef3e96f244493d48e5 /bin/sh -c21.88 MB
97COPY serve-b200.sh /usr/local/bin/serve-b200 # buildkit1.1 KB
98RUN |1 RECIPE_REF=17c58984378af20fcec1f4ef3e96f244493d48e5 /bin/sh -c1.17 KB
99ENTRYPOINT ["/usr/local/bin/serve-b200"]0 B

Command

 ARG RELEASE