Sign inSign up

getlilac/vllm:v0.23.0-glm52-main-parser-mtp-20260619-toolhelper

Manifest digest

sha256:996d155393399e4b18efe45c92061f6900d6f1f8864a6924d643ea1579b4560e

OS/ARCH

linux/amd64

Compressed size

8.57 GB

Last pushed

4 months by themrtoadog

Type

Image

Manifest digest

sha256:996d155393399e4b18efe45c92061f6900d6f1f8864a6924d643ea1579b4560e

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=22.040 B
5ADD file ... in / 29.04 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 -c4.43 MB
13ENV CUDA_VERSION=13.0.20 B
14RUN |1 TARGETARCH=amd64 /bin/sh -c100.3 MB
15RUN |1 TARGETARCH=amd64 /bin/sh -c185 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 DEADSNAKES_MIRROR_URL0 B
24ARG DEADSNAKES_GPGKEY_URL0 B
25ARG GET_PIP_URL0 B
26ENV DEBIAN_FRONTEND=noninteractive0 B
27WORKDIR /vllm-workspace104 B
28RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=230 B
29RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=214.05 MB
30RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=1.61 GB
31RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=48.39 MB
32ENV UV_HTTP_TIMEOUT=5000 B
33ENV UV_INDEX_STRATEGY=unsafe-best-match0 B
34ENV UV_LINK_MODE=copy0 B
35ENV UV_PYTHON_INSTALL_DIR=/opt/uv/python0 B
36ENV UV_CACHE_DIR=/opt/uv/cache0 B
37RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=151 B
38RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=4.8 KB
39COPY docker/entrypoints/vllm-nonroot-entrypoint.sh /usr/local/bin/vllm-nonroot-entrypoint.sh # buildkit2.03 KB
40RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=2.03 KB
41ENV VLLM_ENABLE_CUDA_COMPATIBILITY=00 B
42ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
43COPY requirements/common.txt /tmp/common.txt # buildkit1.58 KB
44COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit648 B
45RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=3.6 GB
46ARG FLASHINFER_VERSION=0.6.120 B
47RUN |7 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=1.44 GB
48ARG GDRCOPY_CUDA_VERSION=12.80 B
49ARG GDRCOPY_OS_VERSION=Ubuntu22_040 B
50ARG TARGETPLATFORM0 B
51COPY tools/install_gdrcopy.sh /tmp/install_gdrcopy.sh # buildkit895 B
52RUN |10 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=524.67 KB
53ARG BITSANDBYTES_VERSION_X86=0.46.10 B
54ARG BITSANDBYTES_VERSION_ARM64=0.42.00 B
55ARG TIMM_VERSION=>=1.0.170 B
56ARG RUNAI_MODEL_STREAMER_VERSION=>=0.15.70 B
57RUN |14 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=115.7 MB
58ARG PIP_INDEX_URL UV_INDEX_URL0 B
59ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
60ARG PYTORCH_CUDA_INDEX_BASE_URL0 B
61ARG PIP_KEYRING_PROVIDER UV_KEYRING_PROVIDER0 B
62ARG PYTORCH_NIGHTLY0 B
63COPY /workspace/torch_lib_versions.txt torch_lib_versions.txt # buildkit198 B
64RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=256.87 MB
65RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=32 B
66ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
67RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=11.04 MB
68RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=76.7 MB
69RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=429.91 MB
70ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
71COPY examples examples # buildkit257.85 KB
72COPY benchmarks benchmarks # buildkit240.82 KB
73COPY ./vllm/collect_env.py . # buildkit9.62 KB
74ARG TARGETPLATFORM0 B
75ARG INSTALL_KV_CONNECTORS=false0 B
76ARG CUDA_VERSION0 B
77ARG VLLM_BUILD_COMMIT0 B
78ARG VLLM_BUILD_PIPELINE0 B
79ARG VLLM_BUILD_URL0 B
80ARG VLLM_IMAGE_TAG0 B
81ARG PIP_INDEX_URL UV_INDEX_URL0 B
82ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
83ENV UV_HTTP_TIMEOUT=5000 B
84ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
85ENV TORCH_CUDA_ARCH_LIST=7.5 8.0 8.6 8.90 B
86RUN |12 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2213.38 MB
87ARG MOONCAKE_WHEEL_AARCH640 B
88ARG MOONCAKE_WHEEL_X86_640 B
89RUN |14 TARGETPLATFORM=linux/amd64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2460.65 MB
90ENV VLLM_USAGE_SOURCE=production-docker-image0 B
91ENV VLLM_BUILD_COMMIT=91df0fad4dc98a67c7659d9dbd915245d5c43d96 VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/2657 VLLM_IMAGE_TAG=vllm/vllm-openai:v0.23.00 B
92LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=91df0fad4dc98a67c7659d9dbd915245d5c43d96 org.opencontainers.image.version=vllm/vllm-openai:v0.23.0 org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/26570 B
93ENTRYPOINT ["vllm" "serve"]0 B
94ARG LILAC_VLLM_REVISION=b049e01e82f7ea19efec79dd3dd08f867d446eaf0 B
95ARG TRANSFORMERS_SPEC=transformers>=5.9.00 B
96LABEL org.opencontainers.image.source=https://github.com/getlilac/vllm0 B
97LABEL org.opencontainers.image.revision=b049e01e82f7ea19efec79dd3dd08f867d446eaf0 B
98LABEL org.opencontainers.image.description=Lilac vLLM 0.23 GLM0 B
99RUN |2 LILAC_VLLM_REVISION=b049e01e82f7ea19efec79dd3dd08f867d446eaf TRANSFORMERS_SPEC=transformers>=5.9.0 /bin/sh32 B
100COPY vllm/entrypoints/openai/chat_completion/serving.py /tmp/lilac-vllm-overlay/vllm/entrypoints/openai/chat_completion/serving.py # buildkit11.79 KB
101COPY vllm/model_executor/layers/attention/mla_attention.py /tmp/lilac-vllm-overlay/vllm/model_executor/layers/attention/mla_attention.py # buildkit20.21 KB
102COPY vllm/model_executor/layers/mla.py /tmp/lilac-vllm-overlay/vllm/model_executor/layers/mla.py # buildkit2.02 KB
103COPY vllm/model_executor/models/deepseek_mtp.py /tmp/lilac-vllm-overlay/vllm/model_executor/models/deepseek_mtp.py # buildkit5.13 KB
104COPY vllm/model_executor/models/deepseek_v2.py /tmp/lilac-vllm-overlay/vllm/model_executor/models/deepseek_v2.py # buildkit14.18 KB
105COPY vllm/parser /tmp/lilac-vllm-overlay/vllm/parser # buildkit32.26 KB
106COPY vllm/reasoning /tmp/lilac-vllm-overlay/vllm/reasoning # buildkit32.79 KB
107COPY vllm/tool_parsers /tmp/lilac-vllm-overlay/vllm/tool_parsers # buildkit131.17 KB
108COPY vllm/v1/attention/backends/mla/flashinfer_mla_sparse.py /tmp/lilac-vllm-overlay/vllm/v1/attention/backends/mla/flashinfer_mla_sparse.py # buildkit3.76 KB
109COPY vllm/v1/attention/backends/mla/flashmla_sparse.py /tmp/lilac-vllm-overlay/vllm/v1/attention/backends/mla/flashmla_sparse.py # buildkit8.3 KB
110COPY vllm/v1/attention/backends/mla/rocm_aiter_mla_sparse.py /tmp/lilac-vllm-overlay/vllm/v1/attention/backends/mla/rocm_aiter_mla_sparse.py # buildkit7.19 KB
111COPY vllm/v1/attention/backends/mla/xpu_mla_sparse.py /tmp/lilac-vllm-overlay/vllm/v1/attention/backends/mla/xpu_mla_sparse.py # buildkit2.74 KB
112COPY vllm/v1/attention/backends/mla/prefill /tmp/lilac-vllm-overlay/vllm/v1/attention/backends/mla/prefill # buildkit8 KB
113COPY vllm/v1/spec_decode/llm_base_proposer.py /tmp/lilac-vllm-overlay/vllm/v1/spec_decode/llm_base_proposer.py # buildkit16.92 KB
114RUN |2 LILAC_VLLM_REVISION=b049e01e82f7ea19efec79dd3dd08f867d446eaf TRANSFORMERS_SPEC=transformers>=5.9.0 /bin/sh14.72 MB

Command

 ARG RELEASE