Sign inSign up

vllm/vime:pr398-90495385-amd64

Index digest

sha256:918a851f2a5e8159d773a6c00d1710b57641d36d9644a9846c6cf7c0d3c33f14

OS/ARCH

linux/amd64

Compressed size

12.27 GB

Last pushed

25 days by vllm

Type

Image

Manifest digest

sha256:637e6728e7253a0a8e21842f1e2deb323f75f6ac853c8c123b0f831305558c28

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 CUDA_VERSION0 B
22ARG PYTHON_VERSION0 B
23ARG NCCL_VERSION0 B
24ARG DEADSNAKES_MIRROR_URL0 B
25ARG DEADSNAKES_GPGKEY_URL0 B
26ARG GET_PIP_URL0 B
27ENV DEBIAN_FRONTEND=noninteractive0 B
28WORKDIR /vllm-workspace104 B
29RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7230 B
30RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7222.51 MB
31RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.71.28 GB
32RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.745.64 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=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7150 B
39RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7162 B
40ENV UV_OVERRIDE=/etc/uv-overrides.txt0 B
41RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.73.62 KB
42COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB
43RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.72.03 KB
44ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B
45ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
46COPY requirements/common.txt /tmp/common.txt # buildkit1.63 KB
47COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit869 B
48RUN |7 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.74.32 GB
49ARG FLASHINFER_VERSION=0.6.16.post30 B
50RUN |8 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.71.4 GB
51ARG GDRCOPY_CUDA_VERSION=12.80 B
52ARG GDRCOPY_OS_VERSION=Ubuntu24_040 B
53ARG TARGETPLATFORM0 B
54COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit896 B
55RUN |11 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.7522.99 KB
56ARG TIMM_VERSION=>=1.0.170 B
57ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
58RUN |13 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 NCCL_VERSION=2.30.758.66 MB
59ARG PIP_INDEX_URL UV_INDEX_URL0 B
60ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
61ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
62ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER0 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 # buildkit239 B
66RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=295.16 MB
67RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=32 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 # buildkit251 B
70RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=13.81 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 # buildkit263.35 KB
73COPY benchmarks benchmarks # buildkit283.22 KB
74COPY ./vllm/collect_env.py . # buildkit9.61 KB
75ARG TARGETPLATFORM0 B
76ARG INSTALL_KV_CONNECTORS=false0 B
77ARG CUDA_VERSION0 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_URL0 B
83ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 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 |12 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2295.11 MB
88ENV VLLM_USAGE_SOURCE=production-docker-image0 B
89ENV VLLM_BUILD_COMMIT=311b3513af33bc29b4acb2fde2e9313e5e9966a0 VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/5250 VLLM_IMAGE_TAG=vllm/vllm-openai:nightly-311b3513af33bc29b4acb2fde2e9313e5e9966a00 B
90LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=311b3513af33bc29b4acb2fde2e9313e5e9966a0 org.opencontainers.image.version=vllm/vllm-openai:nightly-311b3513af33bc29b4acb2fde2e9313e5e9966a0 org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/52500 B
91ENTRYPOINT ["vllm" "serve"]0 B
92ARG PATCH_VERSION=latest0 B
93ARG MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e0870 B
94ARG DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be0 B
95ARG DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a0 B
96ARG DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.30 B
97ARG FLASH_QLA_COMMIT=821fd9d37ede18fdc2a4e707fefe3770bfc32e580 B
98ARG TRANSFORMER_ENGINE_COMMIT=c9877beb87ad7e711e1869dd0b5062167ede447a0 B
99ARG TRANSFORMER_ENGINE_CUDA_ARCHS=90;100a;103a0 B
100ARG TMS_COMMIT=8d30c59ca12a68d9deccbc9c6599076a1218cbc50 B
101ARG ENABLE_CUDA_13=10 B
102ARG FA2_MAX_JOBS=640 B
103WORKDIR /root/32 B
104RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be1.63 GB
105RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be173 B
106RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be81.89 KB
107RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be865.44 MB
108RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be383.32 MB
109RUN |11 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be2.72 MB
110ARG INSTALL_FLASHQLA=00 B
111RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be32 B
112RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be32 B
113RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be239.52 MB
114RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be49.61 MB
115RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be145.28 MB
116RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be1.82 MB
117RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be3.84 MB
118RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be11.72 MB
119RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be114.61 MB
120COPY requirements.txt /tmp/requirements.txt # buildkit437 B
121RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be549.81 MB
122RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be32 B
123RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be113.23 MB
124RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be152 B
125COPY docker/patch/latest/megatron*.patch /root/Megatron-LM/ # buildkit11.17 KB
126RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be135.7 MB
127COPY docker/patch/latest/vllm.patch /tmp/vllm.patch # buildkit9.94 KB
128RUN |12 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be15.51 MB
129ARG VIME_REPO=https://github.com/aoshen02/vime.git0 B
130ARG VIME_COMMIT=c801fa89bd164c65df6158e036b930ffb70fb4980 B
131RUN |14 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be35.29 MB
132RUN |14 PATCH_VERSION=latest MEGATRON_COMMIT=1dcf0dafa884ad52ffb243625717a3471643e087 DEEPGEMM_COMMIT=b38a77cd193cf38f670caae192310521d24343be1.42 MB
133ENTRYPOINT []0 B
134CMD ["/bin/bash"]0 B
135ARG VIME_COMMIT=117cc445e94d5fe2b99f991c0ac599dac868c1e90 B
136LABEL org.opencontainers.image.source=https://github.com/vllm-project/vime org.opencontainers.image.revision=117cc445e94d5fe2b99f991c0ac599dac868c1e90 B
137RUN |1 VIME_COMMIT=117cc445e94d5fe2b99f991c0ac599dac868c1e9 /bin/sh -c117.59 KB
138COPY vllm-mrv2-mimo-kv-table-wake.patch /tmp/mimo-kv-table-wake.patch # buildkit2.83 KB
139RUN /bin/sh -c cd /usr/local/lib/python3.12/dist-packages121.99 KB
140COPY vllm-old.patch /tmp/vllm-old.patch # buildkit9.94 KB
141COPY vllm-pull_weights.patch /tmp/vllm-pull_weights.patch # buildkit4.68 KB
142COPY vllm.patch /tmp/vllm.patch # buildkit2.48 KB
143RUN /bin/sh -c VLLM_SITE="$(python3 -c88.86 KB
144RUN /bin/sh -c cd /root/vime159.49 KB

Command

 ARG RELEASE