Sign inSign up

vllm/vime:pr402-0567e291

Multi-platform
Index digest

sha256:6864c6a7245ea40e226ec101a036632a37f874603c4e47986856af68e08801bc

OS/ARCH

Compressed size

12.3 GB

Last pushed

19 days by vllm

Type

Image

Manifest digest

sha256:fe41cb8dc562f7cede438f5dc7b6580c036884795c1ae582fd09bd86c8c7feff

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=13.0.96-10 B
10ARG TARGETARCH0 B
11LABEL maintainer=NVIDIA CORPORATION <[email protected]>0 B
12RUN |1 TARGETARCH=amd64 /bin/sh -c4.34 MB
13ENV CUDA_VERSION=13.0.20 B
14RUN |1 TARGETARCH=amd64 /bin/sh -c100.29 MB
15RUN |1 TARGETARCH=amd64 /bin/sh -c182 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 INSTALL_RUBIN_PRERELEASE0 B
22ARG CUDA_VERSION0 B
23ARG PYTHON_VERSION0 B
24ARG NCCL_VERSION0 B
25ARG DEADSNAKES_MIRROR_URL0 B
26ARG DEADSNAKES_GPGKEY_URL0 B
27ARG GET_PIP_URL0 B
28ENV CUDA_VERSION=13.0.20 B
29ENV DEBIAN_FRONTEND=noninteractive0 B
30WORKDIR /vllm-workspace104 B
31RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12230 B
32RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12222.51 MB
33RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.121.28 GB
34RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.1232 B
35RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.1238.82 MB
36ENV UV_HTTP_TIMEOUT=5000 B
37ENV UV_INDEX_STRATEGY=unsafe-best-match0 B
38ENV UV_LINK_MODE=copy0 B
39ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python0 B
40ENV UV_CACHE_DIR=/opt/uv/cache0 B
41RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12151 B
42RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12163 B
43ENV UV_OVERRIDE=/etc/uv-overrides.txt0 B
44RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.123.62 KB
45COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB
46RUN |7 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.122.03 KB
47ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B
48ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
49COPY requirements/common.txt /tmp/common.txt # buildkit1.63 KB
50COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit875 B
51RUN |8 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.124.32 GB
52ARG FLASHINFER_VERSION=0.6.170 B
53RUN |9 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.121.41 GB
54ARG GDRCOPY_CUDA_VERSION=12.80 B
55ARG GDRCOPY_OS_VERSION=Ubuntu24_040 B
56ARG TARGETPLATFORM0 B
57COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit896 B
58RUN |12 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12523.07 KB
59ARG TIMM_VERSION=>=1.0.170 B
60ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
61RUN |14 INSTALL_RUBIN_PRERELEASE=false CUDA_VERSION=13.0.2 PYTHON_VERSION=3.1258.72 MB
62ARG PIP_INDEX_URL UV_INDEX_URL0 B
63ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
64ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
65ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER0 B
66ARG PYTORCH_NIGHTLY0 B
67COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit199 B
68COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit242 B
69RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=297.45 MB
70RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=32 B
71ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
72COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit253 B
73RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=13.81 MB
74ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
75COPY examples examples # buildkit268.04 KB
76COPY benchmarks benchmarks # buildkit284.95 KB
77COPY ./vllm/collect_env.py . # buildkit9.61 KB
78ARG TARGETPLATFORM0 B
79ARG INSTALL_KV_CONNECTORS=false0 B
80ARG INSTALL_RUBIN_PRERELEASE0 B
81ARG TRITON_INSTALL_FROM_SOURCE_REPO0 B
82ARG CUDA_VERSION0 B
83ARG VLLM_BUILD_COMMIT0 B
84ARG VLLM_BUILD_PIPELINE0 B
85ARG VLLM_BUILD_URL0 B
86ARG VLLM_IMAGE_TAG0 B
87ARG PIP_INDEX_URL UV_INDEX_URL0 B
88ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
89ENV UV_HTTP_TIMEOUT=5000 B
90ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
91ENV TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 8.90 B
92RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false339.84 MB
93RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false92 B
94COPY /tmp/triton_from_source_workspace/dist/wheel.sha256 /tmp/triton-source-wheel.sha256 # buildkit212 B
95COPY tools/install_triton_from_source.sh /tmp/install_triton_from_source.sh # buildkit823 B
96RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false183 B
97ENV VLLM_USAGE_SOURCE=production-docker-image0 B
98ENV VLLM_BUILD_COMMIT=6f7df92a8e6cdc74a725b8f10b4d0b48ba2b37ef VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/5734 VLLM_IMAGE_TAG=vllm/vllm-openai:nightly-6f7df92a8e6cdc74a725b8f10b4d0b48ba2b37ef0 B
99LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=6f7df92a8e6cdc74a725b8f10b4d0b48ba2b37ef org.opencontainers.image.version=vllm/vllm-openai:nightly-6f7df92a8e6cdc74a725b8f10b4d0b48ba2b37ef org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/57340 B
100ENTRYPOINT ["vllm" "serve"]0 B
101ARG PATCH_VERSION=latest0 B
102ARG MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e0870 B
103ARG DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be0 B
104ARG DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a0 B
105ARG DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.30 B
106ARG FLASH_QLA_COMMIT=821fd9d37ede18fdc2a4e707fefe3770bfc32e580 B
107ARG TRANSFORMER_ENGINE_COMMIT=c9877beb87ad7e711e1869dd0b5062167ede447a0 B
108ARG TRANSFORMER_ENGINE_CUDA_ARCHS=90;100a;103a0 B
109ARG TMS_COMMIT=8d30c59ca12a68d9deccbc9c6599076a1218cbc50 B
110ARG ENABLE_CUDA_13=10 B
111ARG FA2_MAX_JOBS=80 B
112WORKDIR /root/32 B
113RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be1.63 GB
114RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be174 B
115RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be81.89 KB
116RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be865.46 MB
117RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be383.32 MB
118RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be2.72 MB
119ARG INSTALL_FLASHQLA=00 B
120RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be32 B
121RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be32 B
122RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be239.53 MB
123RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be49.61 MB
124RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be133.51 MB
125RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be1.82 MB
126RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be3.87 MB
127RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be11.72 MB
128RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be115.32 MB
129COPY requirements.txt /tmp/requirements.txt # buildkit424 B
130RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be551.56 MB
131RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be32 B
132RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be113.23 MB
133RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be153 B
134COPY docker/patch/latest/megatron*.patch /root/Megatron-LM/ # buildkit11.16 KB
135RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be123.78 MB
136COPY docker/patch/latest/vllm-pull_weights.patch /tmp/vllm-pull_weights.patch # buildkit4.68 KB
137COPY docker/patch/latest/vllm.patch /tmp/vllm.patch # buildkit1.46 KB
138RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be15.45 MB
139ARG VIME_REPO=https://github.com/aoshen02/vime.git0 B
140ARG VIME_COMMIT=0567e291d4b317a104ffa8e85c61b52377637da30 B
141RUN |14 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be34.98 MB
142RUN |14 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be1.42 MB
143ENTRYPOINT []0 B
144CMD ["/bin/bash"]0 B

Command

 ARG RELEASE