Sign inSign up

vllm/vime:vime-slime-commit-540ea919e45b

Multi-platform
Index digest

sha256:cfe7116ae1aa13149e03271e124400e82ac594cd295f48f30ce9576bc48a2889

OS/ARCH

Compressed size

15 GB

Last pushed

9 days by vllm

Type

Image

Manifest digest

sha256:5e0c05795d47b5d482e8619af268d1812b78a791ba00963dc3269baaa63e19e1

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 / 28.55 MB
6CMD ["/bin/bash"]0 B
7ENV NVARCH=sbsa0 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=arm64 /bin/sh -c4.32 MB
13ENV CUDA_VERSION=13.0.20 B
14RUN |1 TARGETARCH=arm64 /bin/sh -c95.11 MB
15RUN |1 TARGETARCH=arm64 /bin/sh -c186 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 TARGETPLATFORM0 B
23ARG max_jobs=20 B
24ARG CUDA_VERSION0 B
25ARG PYTHON_VERSION0 B
26ARG NCCL_VERSION0 B
27ARG DEADSNAKES_MIRROR_URL0 B
28ARG DEADSNAKES_GPGKEY_URL0 B
29ARG GET_PIP_URL0 B
30ENV DEBIAN_FRONTEND=noninteractive0 B
31WORKDIR /vllm-workspace104 B
32RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=16231 B
33RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=16216.5 MB
34RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=161.49 GB
35ENV CUDA_HOME=/usr/local/cuda0 B
36ENV PATH=/usr/local/cuda/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B
37ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
38COPY tools/install_snapshot_runtime.sh /usr/local/bin/install_snapshot_runtime.sh # buildkit1.62 KB
39RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=1632 B
40ENV CRIU_CUDA_PLUGIN_DIR=/usr/local/lib/criu0 B
41RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=1632.57 MB
42ENV UV_HTTP_TIMEOUT=5000 B
43ENV UV_INDEX_STRATEGY=unsafe-best-match0 B
44ENV UV_LINK_MODE=copy0 B
45ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python0 B
46ENV UV_CACHE_DIR=/opt/uv/cache0 B
47ENV UV_COMPILE_BYTECODE=10 B
48RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=16150 B
49RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=16164 B
50ENV UV_OVERRIDE=/etc/uv-overrides.txt0 B
51RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=163.62 KB
52COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB
53RUN |9 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=162.03 KB
54ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B
55ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
56COPY requirements/common.txt /tmp/common.txt # buildkit1.69 KB
57COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit804 B
58RUN |10 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=163.89 GB
59RUN |10 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=162.93 GB
60ARG GDRCOPY_CUDA_VERSION=12.80 B
61ARG GDRCOPY_OS_VERSION=Ubuntu24_040 B
62ARG TARGETPLATFORM0 B
63COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit894 B
64RUN |13 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=16518.72 KB
65ARG TIMM_VERSION=>=1.0.170 B
66ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
67ARG GGUF_VERSION=>=0.19.00 B
68RUN |16 INSTALL_RUBIN_PRERELEASE=false TARGETPLATFORM=linux/arm64 max_jobs=16275.06 MB
69ARG PIP_INDEX_URL UV_INDEX_URL0 B
70ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
71ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
72ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER0 B
73ARG PYTORCH_NIGHTLY0 B
74COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit202 B
75COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit241 B
76RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=544.86 MB
77RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=201 B
78RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=32 B
79ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
80COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit253 B
81RUN |8 PIP_INDEX_URL= UV_INDEX_URL= PIP_EXTRA_INDEX_URL=15.51 MB
82ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
83COPY examples examples # buildkit314.84 KB
84COPY benchmarks benchmarks # buildkit302.03 KB
85COPY ./vllm/collect_env.py . # buildkit9.61 KB
86ARG TARGETPLATFORM0 B
87ARG INSTALL_KV_CONNECTORS=false0 B
88ARG INSTALL_RUBIN_PRERELEASE0 B
89ARG TRITON_INSTALL_FROM_SOURCE_REPO0 B
90ARG CUDA_VERSION0 B
91ARG VLLM_BUILD_COMMIT0 B
92ARG VLLM_BUILD_PIPELINE0 B
93ARG VLLM_BUILD_URL0 B
94ARG VLLM_IMAGE_TAG0 B
95ARG PIP_INDEX_URL UV_INDEX_URL0 B
96ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
97ENV UV_HTTP_TIMEOUT=5000 B
98ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
99ENV TORCH_CUDA_ARCH_LIST=8.0 8.7 8.9 9.00 B
100RUN |14 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false529.74 MB
101RUN |14 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false92 B
102COPY /tmp/triton_from_source_workspace/dist/wheel.sha256 /tmp/triton-source-wheel.sha256 # buildkit212 B
103COPY tools/install_triton_from_source.sh /tmp/install_triton_from_source.sh # buildkit822 B
104RUN |14 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false182 B
105ARG BUILD_NIXL0 B
106RUN |15 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true INSTALL_RUBIN_PRERELEASE=false92 B
107ENV VLLM_USAGE_SOURCE=production-docker-image0 B
108ENV VLLM_BUILD_COMMIT=83c078d955cd6e0c63d3b12abe669c8de936a370 VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/9006 VLLM_IMAGE_TAG=vllm/vllm-openai:nightly-83c078d955cd6e0c63d3b12abe669c8de936a3700 B
109LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=83c078d955cd6e0c63d3b12abe669c8de936a370 org.opencontainers.image.version=vllm/vllm-openai:nightly-83c078d955cd6e0c63d3b12abe669c8de936a370 org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/90060 B
110ENTRYPOINT ["vllm" "serve"]0 B
111ARG PATCH_VERSION=latest0 B
112ARG VLLM_COMMIT=83c078d955cd6e0c63d3b12abe669c8de936a3700 B
113ARG MEGATRON_COMMIT=4b4acac9a1d28ea6829c8d4f566d75698a21249d0 B
114ARG DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b0 B
115ARG DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a0 B
116ARG DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.30 B
117ARG FLASH_QLA_COMMIT=821fd9d37ede18fdc2a4e707fefe3770bfc32e580 B
118ARG TRANSFORMER_ENGINE_VERSION=2.18.00 B
119ARG TRANSFORMER_ENGINE_COMMIT=27486e03cfc1fa41f6932dcecdc47c71c47eac3e0 B
120ARG TRANSFORMER_ENGINE_CUDA_ARCHS=90;100a;103a0 B
121ARG TMS_COMMIT=bdab8c55f490a00a746ec34e301391617f5ca8780 B
122ARG TMS_CUDA_MAJOR=0 B
123ARG NCCL_VERSION=2.30.70 B
124ARG MOONCAKE_VERSION=0.3.120 B
125ARG MAX_JOBS=160 B
126ENV MAX_JOBS=16 CMAKE_BUILD_PARALLEL_LEVEL=160 B
127ENV LD_LIBRARY_PATH=/usr/local/lib/python3.12/dist-packages/nvidia/nccl/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
128ARG ENABLE_CUDA_13=10 B
129ARG FA2_MAX_JOBS=640 B
130WORKDIR /root0 B
131|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1128 B
132|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=11.72 GB
133|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=11.01 KB
134ENV PIP_BREAK_SYSTEM_PACKAGES=10 B
135|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1291 B
136ENV PIP_CONSTRAINT=/opt/vime/constraints.txt0 B
137COPY file:526bfc6171a58f36cc7ecfa912c8f3dc730981428548c39a39c1e2c2a54162c7 in /tmp/flash-attn-cxx20.patch 468 B
138|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1229.79 MB
139|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1378.86 MB
140|16 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=15.03 MB
141ARG INSTALL_FLASHQLA=00 B
142|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=10 B
143|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1362 B
144|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1316.58 MB
145|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=150.13 MB
146|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1183.65 MB
147|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1920.97 KB
148|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=16.64 MB
149|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=111.67 MB
150|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1113.57 MB
151COPY file:4b816d7db31501b9ee307364aa4278dae361096f6fea137b60090253a3e8c309 in /tmp/requirements.txt 465 B
152COPY dir:c0ff781f621ec5e6f3eb365b55f76a98a14078471d3c6ce0ff3d596664e767ad in /tmp/sunabako-wheels/ 859.76 KB
153|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1887.55 KB
154|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1753.79 MB
155|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=144.86 MB
156COPY dir:8f733c581844f5a96c88a5405fd574d39aae71f1b43d543a64af38e153742b09 in /tmp/router-wheels/ 7.84 MB
157|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=18.2 MB
158|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=10 B
159|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=174.22 MB
160|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1155 B
161COPY file:8cfe6685dd1dd786067fc74d97ef1ff166262b92da85fd7a6b8d60ea4f65c389 in /root/Megatron-LM/ 6.63 KB
162|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1167.17 MB
163COPY file:6883111897ee0300f467f887b4f918466f895f88c0c9a61edd257f61ab924bd3 in /tmp/vllm-pull_weights.patch 4.5 KB
164COPY file:87e6abb74f16c413277aabcbcdd0f33d7398d5e33014920238d9e579b494c8ea in /tmp/vllm-score-centering.patch 5.05 KB
165COPY file:2827ec2120985dd83361ecbf05acf5eebefe20996d026bbb0cf8639b7b92a9c1 in /tmp/vllm-pd-request-metrics.patch 6.53 KB
166COPY file:968c7a0803e0f6acc6d5da7bc8d09c4a400b8f3402692269925c991534eb4273 in /tmp/vllm-inflight-queue-diagnostics.patch 3 KB
167COPY file:7d75590fd0751f984bbef4135d283ce39c7556fa61249088087bd42645385007 in /tmp/vllm-generation-health.patch 1.48 KB
168COPY file:2b2069d239c6add769dc52fd3b4653972af9a1988b193bb1846579802e566a2c in /tmp/vllm-tensor-dump.patch 27.51 KB
169COPY file:d0f5a01e30ecae5fa3ac297b40b7a5c304e1fcb94bc4c23ba21413fdf314f37d in /tmp/vllm-fine-grained-tracing.patch 6.22 KB
170COPY file:5696f59a2521c35c37da4dee39d2d1532126c94f5c2d6ff56f4b7d7dc11481d8 in /tmp/vllm-nixl-peer-retirement.patch 3.32 KB
171COPY file:c1be710d277303aeecd3ea5304cf1d4a9d051c06468eef4b0a80ade5f9bd72a3 in /tmp/vllm-host-cache-ratio.patch 720 B
172COPY file:318eb321baebf4a56af230be9a2e217e2fb8e926cbd17969d7cee874a79a53e2 in /tmp/vllm-ec-dp-control.patch 724 B
173COPY file:c0b9695f10d8b94bf9e468143fa88914b6d107dbd3a1adbb443f9398404eef02 in /tmp/vllm-mm-decoder-only.patch 2.08 KB
174COPY file:02a4e49cdc715b19a75078b6f10b68627577bac2285c8efb57030d7b7f27309d in /tmp/vllm-rust-server-info.patch 2.59 KB
175COPY file:c05000d34569d205505ab729631f7e2ff679269034078359accc79635410f689 in /tmp/vllm-deepgemm-bf16.patch 6.85 KB
176COPY file:6baa823943c515bc8becdc9e4a43f5903b2e28625c1310f15100998103e83cac in /tmp/vllm-deepep-optional-shrink.patch 480 B
177COPY file:03351bfd8d677db644b160d510507bbbfdff688f2f2edcc1c6d8cf0b11d07ccf in /tmp/vllm-glm5next-unbounded-gate.patch 872 B
178COPY file:1d0ce89b41368a63cdc3834a95c09a958fdfdeb47d71743ca9045e059f501a5e in /tmp/vllm-utility-rpc-cleanup.patch 2.41 KB
179COPY file:f0112cd9e4b00a6fe45cf805a7b19810cba076be76004d319c221c4fc9c584f5 in /tmp/vllm-tito-columnar-logprobs-render.patch 13.38 KB
180COPY file:a5fb4a37a49f6939de817d0da9d4d38e5cd84223fb943deda6085483bb4089e0 in /tmp/vllm-watchdog-fatal-action.patch 1.52 KB
181COPY file:d73f855d730ef4d0b2f4f6c4e28891044b07ceb392a57af0e9a8c4ec985ff8d6 in /tmp/vllm-hybrid-cache-capacity.patch 75.19 KB
182|17 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=1872.31 KB
183COPY file:c3c3af98af6acb44ca80f2998257386bce9de9e7e2581d969db4d8e192fe72c2 in /usr/local/lib/python3.12/dist-packages/vllm/vllm-rs 22.87 MB
184ARG VIME_REPO=https://github.com/vllm-project/vime.git0 B
185ARG VIME_COMMIT=main0 B
186|19 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=139.14 MB
187|19 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b ENABLE_CUDA_13=11.46 MB
188COPY multi:e561bb65c9566a603ffc57607cf8cfd4ae1bd21f90071f624787ff0fc1a8dd1c in /opt/vime/ 787 B
189ARG CUDA_COMPAT_VERSION=580.178.04-1ubuntu10 B
190|20 CUDA_COMPAT_VERSION=580.178.04-1ubuntu1 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b95.33 MB
191|20 CUDA_COMPAT_VERSION=580.178.04-1ubuntu1 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b215 B
192ENTRYPOINT ["/opt/vime/entrypoint.sh"]0 B
193|20 CUDA_COMPAT_VERSION=580.178.04-1ubuntu1 DEEPEP_COMMIT=6845ffd9d59126ec0030c13e0e155935a61e5b5a DEEPEP_CUDA_ARCH_LIST=9.0;10.0;10.3 DEEPGEMM_COMMIT=8965e4a5cded4041c8dbbe8d35b0be31fd8da90b412.01 MB
194CMD ["/bin/bash"]0 B
195LABEL io.vime.slime-commit=540ea919e45b8b929198abce008a103be9cfc7790 B
196LABEL io.vime.source-commit=0e2815966ba3c58fc1621da7c6faa4f01cd725a00 B
197LABEL io.vime.source-snapshot-sha256=861626cb9edf8c7ea54e52ce603b7e4be84f261250b7c2fc82d103d69c22f9300 B
198LABEL io.vime.vllm-commit=83c078d955cd6e0c63d3b12abe669c8de936a3700 B
199LABEL org.vime.experimental=true0 B
200LABEL org.vime.source-commit=0e2815966ba3c58fc1621da7c6faa4f01cd725a00 B
201LABEL org.vime.source-dirty=false0 B
202ARG VIME_COMMIT0 B
203ARG SOURCE_ARCHIVE_SHA2560 B
204ARG SOURCE_MANIFEST_SHA2560 B
205|3 SOURCE_ARCHIVE_SHA256=92463ad8a7ed834f6821e73a10525d01f0730dc0972b86044189356a45435148 SOURCE_MANIFEST_SHA256=c22e82a38d04e5e6fcff2ebf108a9b69e4c5b139a3a43ed432876b20426ed580 VIME_COMMIT=860417ebc6208450f52de0f781498759cc24b54b /bin/sh6.41 MB
206ADD file ... in /root/vime/ 6.14 MB
207|3 SOURCE_ARCHIVE_SHA256=92463ad8a7ed834f6821e73a10525d01f0730dc0972b86044189356a45435148 SOURCE_MANIFEST_SHA256=c22e82a38d04e5e6fcff2ebf108a9b69e4c5b139a3a43ed432876b20426ed580 VIME_COMMIT=860417ebc6208450f52de0f781498759cc24b54b /bin/sh21.7 KB
208COPY dir:58c6a70ddc044f23efad350a5ef0709222dc81a56413bd8cdbd95a9b7e917d82 in /usr/local/lib/python3.12/dist-packages/ 8.23 MB
209COPY dir:e6adc82002884bb6fd24ca920a33a4ac2ea7aaa35d7a210ba4b5cdcc53aaceef in /tmp/router-final-wheels/ 7.84 MB
210|3 SOURCE_ARCHIVE_SHA256=92463ad8a7ed834f6821e73a10525d01f0730dc0972b86044189356a45435148 SOURCE_MANIFEST_SHA256=c22e82a38d04e5e6fcff2ebf108a9b69e4c5b139a3a43ed432876b20426ed580 VIME_COMMIT=860417ebc6208450f52de0f781498759cc24b54b /bin/sh16.37 MB
211COPY file:f103f9bb149639296e5d3260189a4bbc9c6f052e4bf3949cb13651b7a8cadee3 in /tmp/vime-source/source.sha256 32.99 KB
212COPY file:eef40d0b0213d3bf554beb9607df90c79fccd820f85ba0b7291e8dec2d256c14 in /tmp/vime-source/runtime.sha256 734 B
213|3 SOURCE_ARCHIVE_SHA256=92463ad8a7ed834f6821e73a10525d01f0730dc0972b86044189356a45435148 SOURCE_MANIFEST_SHA256=c22e82a38d04e5e6fcff2ebf108a9b69e4c5b139a3a43ed432876b20426ed580 VIME_COMMIT=860417ebc6208450f52de0f781498759cc24b54b /bin/sh0 B
214LABEL io.vime.source-commit=860417ebc6208450f52de0f781498759cc24b54b org.vime.source-commit=860417ebc6208450f52de0f781498759cc24b54b org.vime.source-dirty=false0 B
215ARG VIME_COMMIT=84a012353b689dc086c1f425859b3f5e692a95420 B
216RUN |1 VIME_COMMIT=84a012353b689dc086c1f425859b3f5e692a9542 /bin/sh -c274 B
217COPY /usr/local/lib/python3.12/dist-packages/vllm/ /usr/local/lib/python3.12/dist-packages/vllm/ # buildkit333 MB
218ADD engine.tar /usr/local/lib/python3.12/dist-packages/ # buildkit9.53 MB
219COPY /wheels/ /tmp/vime-router-wheels/ # buildkit7.84 MB
220RUN |1 VIME_COMMIT=84a012353b689dc086c1f425859b3f5e692a9542 /bin/sh -c8.2 MB
221RUN |1 VIME_COMMIT=84a012353b689dc086c1f425859b3f5e692a9542 /bin/sh -c12.24 MB
222LABEL io.vime.source-commit=84a012353b689dc086c1f425859b3f5e692a95420 B

Command

 ARG RELEASE