Sign inSign up

ardge/mirrored-llama.cpp:server-cuda13-b12033

Multi-platform
Index digest

sha256:38ca42e8fcc5e1de7759131069683cff636c159c428a1cd7b9c7eeae5507a2f5

OS/ARCH

Compressed size

2.38 GB

Last pushed

1 day by ardge

Type

Image

Manifest digest

sha256:029ae8a7f24515d1d25feeb2bdb317568259e1cb3e8b2c9c09b5eb477880e6be

Image Layers

Image Layers

1ARG RELEASE0 B
2ARG LAUNCHPAD_BUILD_ARCH0 B
3LABEL org.opencontainers.image.version=24.040 B
4ADD file ... in / 27.6 MB
5CMD ["/bin/bash"]0 B
6ENV NVARCH=sbsa0 B
7ENV NVIDIA_REQUIRE_CUDA=cuda>=13.4 brand=unknown,driver>=535,driver<536 brand=grid,driver>=535,driver<536 brand=tesla,driver>=535,driver<5360 B
8ENV NV_CUDA_CUDART_VERSION=13.4.49-10 B
9ARG TARGETARCH0 B
10LABEL maintainer=NVIDIA CORPORATION <[email protected]>0 B
11ARG IMAGE_UPGRADE_ID=00 B
12RUN |2 TARGETARCH=arm64 IMAGE_UPGRADE_ID=20260914 /bin/sh9.9 MB
13ENV CUDA_VERSION=13.4.10 B
14RUN |2 TARGETARCH=arm64 IMAGE_UPGRADE_ID=20260914 /bin/sh398.13 KB
15RUN |2 TARGETARCH=arm64 IMAGE_UPGRADE_ID=20260914 /bin/sh153.01 MB
16RUN |2 TARGETARCH=arm64 IMAGE_UPGRADE_ID=20260914 /bin/sh152.92 MB
17RUN |2 TARGETARCH=arm64 IMAGE_UPGRADE_ID=20260914 /bin/sh187 B
18ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin0 B
19ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib640 B
20COPY NGC-DL-CONTAINER-LICENSE / # buildkit6.72 KB
21ENV NVIDIA_VISIBLE_DEVICES=all0 B
22ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility0 B
23ENV NV_CUDA_LIB_VERSION=13.4.1-10 B
24ENV NV_NVTX_VERSION=13.4.49-10 B
25ENV NV_LIBNPP_VERSION=13.2.0.35-10 B
26ENV NV_LIBNPP_PACKAGE=libnpp-13-4=13.2.0.35-10 B
27ENV NV_LIBCUSPARSE_VERSION=12.8.6.49-10 B
28ENV NV_LIBCUBLAS_PACKAGE_NAME=libcublas-13-40 B
29ENV NV_LIBCUBLAS_VERSION=13.7.0.27-10 B
30ENV NV_LIBCUBLAS_PACKAGE=libcublas-13-4=13.7.0.27-10 B
31ENV NV_LIBNCCL_PACKAGE_NAME=libnccl20 B
32ENV NV_LIBNCCL_PACKAGE_VERSION=2.31.2-10 B
33ENV NCCL_VERSION=2.31.2-10 B
34ENV NV_LIBNCCL_PACKAGE=libnccl2=2.31.2-1+cuda13.40 B
35ARG TARGETARCH0 B
36LABEL maintainer=NVIDIA CORPORATION <[email protected]>0 B
37RUN |1 TARGETARCH=arm64 /bin/sh -c1.66 GB
38RUN |1 TARGETARCH=arm64 /bin/sh -c59.1 KB
39COPY entrypoint.d/ /opt/nvidia/entrypoint.d/ # buildkit1.64 KB
40COPY nvidia_entrypoint.sh /opt/nvidia/ # buildkit1.49 KB
41ENV NVIDIA_PRODUCT_NAME=CUDA0 B
42ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]0 B
43ARG BUILD_DATE=2026-11-02T04:25:10Z0 B
44ARG APP_VERSION=b120330 B
45ARG APP_REVISION=785322d17f2d51e5185cf996a1f4e46d876408410 B
46ARG IMAGE_URL=https://github.com/ggml-org/llama.cpp0 B
47ARG IMAGE_SOURCE=https://github.com/ggml-org/llama.cpp0 B
48LABEL org.opencontainers.image.created=2026-11-02T04:25:10Z org.opencontainers.image.version=b12033 org.opencontainers.image.revision=785322d17f2d51e5185cf996a1f4e46d87640841 org.opencontainers.image.title=llama.cpp0 B
49RUN |5 BUILD_DATE=2026-11-02T04:25:10Z APP_VERSION=b12033 APP_REVISION=785322d17f2d51e5185cf996a1f4e46d87640841240.07 MB
50COPY /app/lib/ /app # buildkit148.24 MB
51ENV LLAMA_ARG_HOST=0.0.0.00 B
52ENV LLAMA_ARG_PORT=80800 B
53COPY /app/full/llama /app/full/llama-server /app/ #26.83 KB
54WORKDIR /app32 B
55HEALTHCHECK {Test:[CMD curl -f http://localhost:8080/health]0 B
56ENTRYPOINT ["/app/llama-server"]0 B

Command

 ARG RELEASE