Docker拉取命令

docker pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64
docker tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64  docker.io/hicasper/llama-cpp-cuda11:server-b10068

Containerd拉取命令

ctr images pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64
ctr images tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64  docker.io/hicasper/llama-cpp-cuda11:server-b10068

Shell快速替换命令

sed -i 's#hicasper/llama-cpp-cuda11:server-b10068#swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64#' deployment.yaml

Ansible快速分发-Docker

#ansible k8s -m shell -a 'docker pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64 && docker tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64  docker.io/hicasper/llama-cpp-cuda11:server-b10068'

Ansible快速分发-Containerd

#ansible k8s -m shell -a 'ctr images pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64 && ctr images tag  swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64  docker.io/hicasper/llama-cpp-cuda11:server-b10068'

镜像构建历史

# 2026-07-19 22:10:16  0.00B 配置容器启动时运行的命令
ENTRYPOINT ["/app/llama-server"]
                        
# 2026-07-19 22:10:16  0.00B 指定检查容器健康状态的命令
HEALTHCHECK {Test:[CMD curl -f http://localhost:8080/health] Interval:0s Timeout:0s StartPeriod:0s StartInterval:0s Retries:0}
                        
# 2026-07-19 22:10:16  0.00B 设置工作目录为/app
WORKDIR /app
                        
# 2026-07-19 22:10:16  71.78KB 复制新文件或目录到容器中
COPY /app/full/llama /app/full/llama-server /app # buildkit
                        
# 2026-07-19 22:10:16  0.00B 设置环境变量 LLAMA_ARG_HOST
ENV LLAMA_ARG_HOST=0.0.0.0
                        
# 2026-07-19 22:10:16  162.07MB 复制新文件或目录到容器中
COPY /app/lib/ /app # buildkit
                        
# 2026-07-19 22:01:05  338.55MB 执行命令并创建新的镜像层
RUN |5 BUILD_DATE=2026-07-19T13:58:55Z APP_VERSION=b10068 APP_REVISION= IMAGE_URL=https://github.com/ggml-org/llama.cpp IMAGE_SOURCE=https://github.com/ggml-org/llama.cpp /bin/sh -c apt-get update     && apt-get install -y libgomp1 curl ffmpeg     && apt autoremove -y     && apt clean -y     && rm -rf /tmp/* /var/tmp/*     && find /var/cache/apt/archives /var/lib/apt/lists -not -name lock -type f -delete     && find /var/cache -type f -delete # buildkit
                        
# 2026-07-19 22:01:05  0.00B 添加元数据标签
LABEL org.opencontainers.image.created=2026-07-19T13:58:55Z org.opencontainers.image.version=b10068 org.opencontainers.image.title=llama.cpp org.opencontainers.image.description=LLM inference in C/C++ org.opencontainers.image.url=https://github.com/ggml-org/llama.cpp org.opencontainers.image.source=https://github.com/ggml-org/llama.cpp
                        
# 2026-07-19 22:01:05  0.00B 定义构建参数
ARG IMAGE_SOURCE=https://github.com/ggml-org/llama.cpp
                        
# 2026-07-19 22:01:05  0.00B 定义构建参数
ARG IMAGE_URL=https://github.com/ggml-org/llama.cpp
                        
# 2026-07-19 22:01:05  0.00B 定义构建参数
ARG APP_REVISION=
                        
# 2026-07-19 22:01:05  0.00B 定义构建参数
ARG APP_VERSION=b10068
                        
# 2026-07-19 22:01:05  0.00B 定义构建参数
ARG BUILD_DATE=2026-07-19T13:58:55Z
                        
# 2023-11-10 16:19:17  0.00B 配置容器启动时运行的命令
ENTRYPOINT ["/opt/nvidia/nvidia_entrypoint.sh"]
                        
# 2023-11-10 16:19:17  0.00B 设置环境变量 NVIDIA_PRODUCT_NAME
ENV NVIDIA_PRODUCT_NAME=CUDA
                        
# 2023-11-10 16:19:17  2.53KB 复制新文件或目录到容器中
COPY nvidia_entrypoint.sh /opt/nvidia/ # buildkit
                        
# 2023-11-10 16:19:17  3.06KB 复制新文件或目录到容器中
COPY entrypoint.d/ /opt/nvidia/entrypoint.d/ # buildkit
                        
# 2023-11-10 16:19:17  256.05KB 执行命令并创建新的镜像层
RUN |1 TARGETARCH=arm64 /bin/sh -c apt-mark hold ${NV_LIBCUBLAS_PACKAGE_NAME} ${NV_LIBNCCL_PACKAGE_NAME} # buildkit
                        
# 2023-11-10 16:19:15  2.03GB 执行命令并创建新的镜像层
RUN |1 TARGETARCH=arm64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends     cuda-libraries-11-4=${NV_CUDA_LIB_VERSION}     ${NV_LIBNPP_PACKAGE}     cuda-nvtx-11-4=${NV_NVTX_VERSION}     libcusparse-11-4=${NV_LIBCUSPARSE_VERSION}     ${NV_LIBCUBLAS_PACKAGE}     ${NV_LIBNCCL_PACKAGE}     && rm -rf /var/lib/apt/lists/* # buildkit
                        
# 2023-11-10 16:19:15  0.00B 添加元数据标签
LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>
                        
# 2023-11-10 16:19:15  0.00B 定义构建参数
ARG TARGETARCH
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBNCCL_PACKAGE
ENV NV_LIBNCCL_PACKAGE=libnccl2=2.11.4-1+cuda11.4
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NCCL_VERSION
ENV NCCL_VERSION=2.11.4-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBNCCL_PACKAGE_VERSION
ENV NV_LIBNCCL_PACKAGE_VERSION=2.11.4-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBNCCL_PACKAGE_NAME
ENV NV_LIBNCCL_PACKAGE_NAME=libnccl2
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBCUBLAS_PACKAGE
ENV NV_LIBCUBLAS_PACKAGE=libcublas-11-4=11.6.5.2-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBCUBLAS_VERSION
ENV NV_LIBCUBLAS_VERSION=11.6.5.2-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBCUBLAS_PACKAGE_NAME
ENV NV_LIBCUBLAS_PACKAGE_NAME=libcublas-11-4
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBCUSPARSE_VERSION
ENV NV_LIBCUSPARSE_VERSION=11.6.0.120-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBNPP_PACKAGE
ENV NV_LIBNPP_PACKAGE=libnpp-11-4=11.4.0.110-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_LIBNPP_VERSION
ENV NV_LIBNPP_VERSION=11.4.0.110-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_NVTX_VERSION
ENV NV_NVTX_VERSION=11.4.120-1
                        
# 2023-11-10 16:19:15  0.00B 设置环境变量 NV_CUDA_LIB_VERSION
ENV NV_CUDA_LIB_VERSION=11.4.3-1
                        
# 2023-11-10 16:12:21  0.00B 设置环境变量 NVIDIA_DRIVER_CAPABILITIES
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
                        
# 2023-11-10 16:12:21  0.00B 设置环境变量 NVIDIA_VISIBLE_DEVICES
ENV NVIDIA_VISIBLE_DEVICES=all
                        
# 2023-11-10 16:12:21  17.29KB 复制新文件或目录到容器中
COPY NGC-DL-CONTAINER-LICENSE / # buildkit
                        
# 2023-11-10 16:12:21  0.00B 设置环境变量 LD_LIBRARY_PATH
ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
                        
# 2023-11-10 16:12:21  0.00B 设置环境变量 PATH
ENV PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
                        
# 2023-11-10 16:12:21  46.00B 执行命令并创建新的镜像层
RUN |1 TARGETARCH=arm64 /bin/sh -c echo "/usr/local/nvidia/lib" >> /etc/ld.so.conf.d/nvidia.conf     && echo "/usr/local/nvidia/lib64" >> /etc/ld.so.conf.d/nvidia.conf # buildkit
                        
# 2023-11-10 16:12:20  1.42MB 执行命令并创建新的镜像层
RUN |1 TARGETARCH=arm64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends     cuda-cudart-11-4=${NV_CUDA_CUDART_VERSION}     ${NV_CUDA_COMPAT_PACKAGE}     && rm -rf /var/lib/apt/lists/* # buildkit
                        
# 2023-11-10 16:11:18  0.00B 设置环境变量 CUDA_VERSION
ENV CUDA_VERSION=11.4.3
                        
# 2023-11-10 16:11:18  17.42MB 执行命令并创建新的镜像层
RUN |1 TARGETARCH=arm64 /bin/sh -c apt-get update && apt-get install -y --no-install-recommends     gnupg2 curl ca-certificates &&     curl -fsSL https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/${NVARCH}/3bf863cc.pub | apt-key add - &&     echo "deb https://developer.download.nvidia.com/compute/cuda/repos/ubuntu2004/${NVARCH} /" > /etc/apt/sources.list.d/cuda.list &&     apt-get purge --autoremove -y curl     && rm -rf /var/lib/apt/lists/* # buildkit
                        
# 2023-11-10 16:11:18  0.00B 添加元数据标签
LABEL maintainer=NVIDIA CORPORATION <cudatools@nvidia.com>
                        
# 2023-11-10 16:11:18  0.00B 定义构建参数
ARG TARGETARCH
                        
# 2023-11-10 16:11:18  0.00B 设置环境变量 NV_CUDA_CUDART_VERSION
ENV NV_CUDA_CUDART_VERSION=11.4.148-1
                        
# 2023-11-10 16:11:18  0.00B 设置环境变量 NVIDIA_REQUIRE_CUDA
ENV NVIDIA_REQUIRE_CUDA=cuda>=11.4
                        
# 2023-11-10 16:11:18  0.00B 设置环境变量 NVARCH
ENV NVARCH=sbsa
                        
# 2023-10-03 19:04:17  0.00B 
/bin/sh -c #(nop)  CMD ["/bin/bash"]
                        
# 2023-10-03 19:04:16  65.68MB 
/bin/sh -c #(nop) ADD file:f70cc2610ea8fcd25e6e9ae727eb9345d5b7198102f6a6d8e458ab8f99efefc3 in / 
                        
# 2023-10-03 19:04:10  0.00B 
/bin/sh -c #(nop)  LABEL org.opencontainers.image.version=20.04
                        
# 2023-10-03 19:04:09  0.00B 
/bin/sh -c #(nop)  LABEL org.opencontainers.image.ref.name=ubuntu
                        
# 2023-10-03 19:04:09  0.00B 
/bin/sh -c #(nop)  ARG LAUNCHPAD_BUILD_ARCH
                        
# 2023-10-03 19:04:09  0.00B 
/bin/sh -c #(nop)  ARG RELEASE
                        
                    

镜像信息

{
    "Id": "sha256:f5e604e218a9db9ca50b809da898e7e294a788efa4d3f7bb6a17972274b1c605",
    "RepoTags": [
        "hicasper/llama-cpp-cuda11:server-b10068",
        "swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11:server-b10068-linuxarm64"
    ],
    "RepoDigests": [
        "hicasper/llama-cpp-cuda11@sha256:d555c64b9809698e7ac7181a00e8b3418c6a4d63dc0c4cc7e5d9109ee3dd54b1",
        "swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/hicasper/llama-cpp-cuda11@sha256:6330fc1201290fbd3e2cadfaa97f2d2fd265acc84679bd3c21435a7b0e7e51b1"
    ],
    "Parent": "",
    "Comment": "buildkit.dockerfile.v0",
    "Created": "2026-07-19T14:10:16.325027375Z",
    "Container": "",
    "ContainerConfig": null,
    "DockerVersion": "",
    "Author": "",
    "Config": {
        "Hostname": "",
        "Domainname": "",
        "User": "",
        "AttachStdin": false,
        "AttachStdout": false,
        "AttachStderr": false,
        "Tty": false,
        "OpenStdin": false,
        "StdinOnce": false,
        "Env": [
            "PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin",
            "NVARCH=sbsa",
            "NVIDIA_REQUIRE_CUDA=cuda\u003e=11.4",
            "NV_CUDA_CUDART_VERSION=11.4.148-1",
            "CUDA_VERSION=11.4.3",
            "LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64",
            "NVIDIA_VISIBLE_DEVICES=all",
            "NVIDIA_DRIVER_CAPABILITIES=compute,utility",
            "NV_CUDA_LIB_VERSION=11.4.3-1",
            "NV_NVTX_VERSION=11.4.120-1",
            "NV_LIBNPP_VERSION=11.4.0.110-1",
            "NV_LIBNPP_PACKAGE=libnpp-11-4=11.4.0.110-1",
            "NV_LIBCUSPARSE_VERSION=11.6.0.120-1",
            "NV_LIBCUBLAS_PACKAGE_NAME=libcublas-11-4",
            "NV_LIBCUBLAS_VERSION=11.6.5.2-1",
            "NV_LIBCUBLAS_PACKAGE=libcublas-11-4=11.6.5.2-1",
            "NV_LIBNCCL_PACKAGE_NAME=libnccl2",
            "NV_LIBNCCL_PACKAGE_VERSION=2.11.4-1",
            "NCCL_VERSION=2.11.4-1",
            "NV_LIBNCCL_PACKAGE=libnccl2=2.11.4-1+cuda11.4",
            "NVIDIA_PRODUCT_NAME=CUDA",
            "LLAMA_ARG_HOST=0.0.0.0"
        ],
        "Cmd": null,
        "Healthcheck": {
            "Test": [
                "CMD",
                "curl",
                "-f",
                "http://localhost:8080/health"
            ]
        },
        "Image": "",
        "Volumes": null,
        "WorkingDir": "/app",
        "Entrypoint": [
            "/app/llama-server"
        ],
        "OnBuild": null,
        "Labels": {
            "maintainer": "NVIDIA CORPORATION \u003ccudatools@nvidia.com\u003e",
            "org.opencontainers.image.created": "2026-07-19T13:58:55Z",
            "org.opencontainers.image.description": "LLM inference in C/C++",
            "org.opencontainers.image.ref.name": "ubuntu",
            "org.opencontainers.image.source": "https://github.com/ggml-org/llama.cpp",
            "org.opencontainers.image.title": "llama.cpp",
            "org.opencontainers.image.url": "https://github.com/ggml-org/llama.cpp",
            "org.opencontainers.image.version": "b10068"
        }
    },
    "Architecture": "arm64",
    "Os": "linux",
    "Size": 2616140109,
    "GraphDriver": {
        "Data": {
            "LowerDir": "/var/lib/docker/overlay2/fb642081a4fc293fcf47ce1eb8e616b41f9ccfc081cbb7cdaaf4615edac3ab54/diff:/var/lib/docker/overlay2/bf11112a99d1755670dc5ec9fa70bcc99b12532f950d1f04e8b7cb392b1fe789/diff:/var/lib/docker/overlay2/4fa8816e4a6f3e88f8c07dd3a47d4f2fd975e0338b96fed09363a9fe01dc5ab2/diff:/var/lib/docker/overlay2/e44152935af803ff85d4923cd513752fdbe2e240e0cd9940a94cae614f4ae7c7/diff:/var/lib/docker/overlay2/fe306cfaf20aae06b9814e572bf98f455dc3d9350cc023f9f684c782aac1f271/diff:/var/lib/docker/overlay2/8729aa7858a3adc9964c27f353917011a3b7cbb7334634c03142d477446f28d6/diff:/var/lib/docker/overlay2/981f9323e107ed29f7ac22b4e2a5ff3b3bf0081404d969c40f57401caf7279a4/diff:/var/lib/docker/overlay2/eeba33cc131b2c2bc981bda8348a0d9d427addfaf63e48385cd6af10cd60dd47/diff:/var/lib/docker/overlay2/4fd916c1ab17b30e00e1ffc8562cbc2d15c2e21ca5be86acf11c571f73aab9ed/diff:/var/lib/docker/overlay2/24faa8a2f363edfeaf5b9aab9b8ad682fc7b61b057a32dc81acd48881a8b0cf4/diff:/var/lib/docker/overlay2/29bd0343e1fc1190565f3bfddf96e95cfd60e0e9031dddf9bedec6a11136b4b2/diff:/var/lib/docker/overlay2/6f5304663e01de9584e7427a4bc099acf3d0f1aa4acf43a14bad88a6012285f2/diff",
            "MergedDir": "/var/lib/docker/overlay2/93da6b8557da3a21c41fa941ba4f3575944a2355fb5be29b2aba4ed953efe696/merged",
            "UpperDir": "/var/lib/docker/overlay2/93da6b8557da3a21c41fa941ba4f3575944a2355fb5be29b2aba4ed953efe696/diff",
            "WorkDir": "/var/lib/docker/overlay2/93da6b8557da3a21c41fa941ba4f3575944a2355fb5be29b2aba4ed953efe696/work"
        },
        "Name": "overlay2"
    },
    "RootFS": {
        "Type": "layers",
        "Layers": [
            "sha256:701ccb11c0525a289459678e1ef3e9269b62737fafb033db91061ae53a0e9c1e",
            "sha256:8fa84dcb0a0806026522fc045de928e89895f938fe193654c5c05ea292140dd2",
            "sha256:4a3af3f5ce449022d5b805003125969fc1116eed548b4c8831a15f664d540374",
            "sha256:486c8d171d8816e5d53b23fbb89a169ae6ca17b005325dfa05ad3b7917f7e79c",
            "sha256:7265cab6e2b83ec0d3fec7e899ee5cc4919ee06ff01ad7aaaa7599f0ecee6c01",
            "sha256:032ccaec9ba9f58debf09b0ce23af95cef7763754046730ea441f6add18c8791",
            "sha256:995a13566c50bdf16c7aa6eea6a52531e643087783339c1fd5437a9fc5f63891",
            "sha256:f055f5f7f24143f7c3e59d9c779908251ec84fab91a087d0a0084cba7029df6e",
            "sha256:75da8f0eda1c92537eeec7860e5eb39cfc993b7d72e6a1b660871c088abb9cee",
            "sha256:fcf52487bde489c18a303573a70f317acf9cba0e50c93586d4be92b1d84dd630",
            "sha256:564f18754e493888dd6a0ad9135e022d7717d6b150cdcd4d952ecf08a8d92279",
            "sha256:1602f8f88633258c21bd97fc3ed5b8ca5bb0ad4aff5e7fa6667a5011a01fd902",
            "sha256:5f70bf18a086007016e948b04aed3b82103a36bea41755b6cddfaf10ace3c6ef"
        ]
    },
    "Metadata": {
        "LastTagTime": "2026-07-19T22:40:05.265162342+08:00"
    }
}