Sign inSign up

getlilac/vllm:v0.24.0-glm52-20260707

Index digest

sha256:fa12754ec7542ac8d3dc6fb463d7b5fbcf97a5575349a97aba456512ce54c672

OS/ARCH

linux/arm64

Compressed size

9.91 GB

Last pushed

2 months by getlilac

Type

Image

Manifest digest

sha256:e007a4273db318435e8a15905da2f69896944f1029c1cb4c11949f229de2d36c

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 / 27.11 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 -c4.38 MB
13ENV CUDA_VERSION=13.0.20 B
14RUN |1 TARGETARCH=arm64 /bin/sh -c95.12 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 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=229 B
29RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=201.3 MB
30RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=1.78 GB
31RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=48.51 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=150 B
38RUN |5 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=4.81 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 # buildkit1.64 KB
44COPY requirements/cuda.txt /tmp/requirements-cuda.txt # buildkit646 B
45RUN |6 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=3.66 GB
46ARG FLASHINFER_VERSION=0.6.120 B
47RUN |7 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=1.63 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 # buildkit896 B
52RUN |10 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=521.05 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=300.17 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
64COPY /workspace/dist/wheel.sha256 /tmp/vllm-wheel.sha256 # buildkit227 B
65RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=464.53 MB
66RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=32 B
67ENV LD_LIBRARY_PATH=/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
68COPY /tmp/ep_kernels_workspace/dist/wheels.sha256 /tmp/ep-kernels-wheels.sha256 # buildkit253 B
69RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=10.8 MB
70RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=78.79 MB
71RUN |22 CUDA_VERSION=13.0.2 PYTHON_VERSION=3.12 DEADSNAKES_MIRROR_URL=429.91 MB
72ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
73COPY examples examples # buildkit257.85 KB
74COPY benchmarks benchmarks # buildkit249.62 KB
75COPY ./vllm/collect_env.py . # buildkit9.62 KB
76ARG TARGETPLATFORM0 B
77ARG INSTALL_KV_CONNECTORS=false0 B
78ARG CUDA_VERSION0 B
79ARG VLLM_BUILD_COMMIT0 B
80ARG VLLM_BUILD_PIPELINE0 B
81ARG VLLM_BUILD_URL0 B
82ARG VLLM_IMAGE_TAG0 B
83ARG PIP_INDEX_URL UV_INDEX_URL0 B
84ARG PIP_EXTRA_INDEX_URL UV_EXTRA_INDEX_URL0 B
85ENV UV_HTTP_TIMEOUT=5000 B
86ARG torch_cuda_arch_list=7.5 8.0 8.6 8.90 B
87ENV TORCH_CUDA_ARCH_LIST=8.0 8.7 8.9 9.00 B
88RUN |12 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2771.28 MB
89ARG MOONCAKE_WHEEL_AARCH640 B
90ARG MOONCAKE_WHEEL_X86_640 B
91RUN |14 TARGETPLATFORM=linux/arm64 INSTALL_KV_CONNECTORS=true CUDA_VERSION=13.0.2446.18 MB
92ENV VLLM_USAGE_SOURCE=production-docker-image0 B
93ENV VLLM_BUILD_COMMIT=ee0da84ab9e04ac7610e28580af62c365e898389 VLLM_BUILD_PIPELINE=019d130e-464e-4ff7-b84b-492992c0c06b VLLM_BUILD_URL=https://buildkite.com/vllm/release-v2/builds/3274 VLLM_IMAGE_TAG=vllm/vllm-openai:v0.24.00 B
94LABEL org.opencontainers.image.source=https://github.com/vllm-project/vllm org.opencontainers.image.revision=ee0da84ab9e04ac7610e28580af62c365e898389 org.opencontainers.image.version=vllm/vllm-openai:v0.24.0 org.opencontainers.image.url=https://buildkite.com/vllm/release-v2/builds/32740 B
95ENTRYPOINT ["vllm" "serve"]0 B
96ARG LILAC_VLLM_REVISION=lilac/v0.24.0-glm520 B
97ARG TRANSFORMERS_SPEC=transformers>=5.9.00 B
98LABEL org.opencontainers.image.source=https://github.com/getlilac/vllm0 B
99LABEL org.opencontainers.image.revision=lilac/v0.24.0-glm520 B
100LABEL org.opencontainers.image.description=Lilac vLLM v0.24.0 GLM0 B
101RUN |2 LILAC_VLLM_REVISION=lilac/v0.24.0-glm52 TRANSFORMERS_SPEC=transformers>=5.9.0 /bin/sh32 B
102COPY vllm/entrypoints/openai/chat_completion/serving.py /tmp/lilac-vllm-overlay/vllm/entrypoints/openai/chat_completion/serving.py # buildkit9.87 KB
103COPY vllm/model_executor/layers/fused_moe/router/gate_linear.py /tmp/lilac-vllm-overlay/vllm/model_executor/layers/fused_moe/router/gate_linear.py # buildkit2.42 KB
104COPY vllm/model_executor/models/deepseek_v2.py /tmp/lilac-vllm-overlay/vllm/model_executor/models/deepseek_v2.py # buildkit14.35 KB
105COPY vllm/parser/abstract_parser.py /tmp/lilac-vllm-overlay/vllm/parser/abstract_parser.py # buildkit6.24 KB
106COPY vllm/parser/engine/parser_engine.py /tmp/lilac-vllm-overlay/vllm/parser/engine/parser_engine.py # buildkit7.97 KB
107COPY vllm/parser/engine/parser_engine_config.py /tmp/lilac-vllm-overlay/vllm/parser/engine/parser_engine_config.py # buildkit1.63 KB
108COPY vllm/parser/engine/streaming_parser_engine.py /tmp/lilac-vllm-overlay/vllm/parser/engine/streaming_parser_engine.py # buildkit3.89 KB
109COPY vllm/parser/engine/token_id_scanner.py /tmp/lilac-vllm-overlay/vllm/parser/engine/token_id_scanner.py # buildkit3.21 KB
110COPY vllm/parser/gemma4.py /tmp/lilac-vllm-overlay/vllm/parser/gemma4.py # buildkit4.7 KB
111RUN |2 LILAC_VLLM_REVISION=lilac/v0.24.0-glm52 TRANSFORMERS_SPEC=transformers>=5.9.0 /bin/sh14.69 MB
112RUN |2 LILAC_VLLM_REVISION=lilac/v0.24.0-glm52 TRANSFORMERS_SPEC=transformers>=5.9.0 /bin/sh9.52 MB

Command

 ARG RELEASE