starred/ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-26 10:45:57 +02:00
Code Issues Packages Projects Releases Wiki Activity
Files
8300a55e1d18f12597fd37dbe6bc57f679d0ceda
ollama/Dockerfile
Masato Nakasaka af5f5bdf60 Removed libcap related code 
libcap is not directly related to Vulkan and should be added by its own PR. It adds additional library dependencies for building and also requires users to run setcap or run ollama as root, which is not ideal for easy use
2025-08-27 11:51:53 +09:00

142 lines
6.2 KiB
Docker
Raw Blame History

# vim: filetype=dockerfile
ARG FLAVOR=${TARGETARCH}
ARG ROCMVERSION=6.3.3
ARG JETPACK5VERSION=r35.4.1
ARG JETPACK6VERSION=r36.4.0
ARG CMAKEVERSION=3.31.2
ARG VULKANVERSION=1.4.304.1
# We require gcc v10 minimum. v10.3 has regressions, so the rockylinux 8.5 AppStream has the latest compatible version
FROM --platform=linux/amd64 rocm/dev-almalinux-8:${ROCMVERSION}-complete AS base-amd64
RUN yum install -y yum-utils \
&& yum-config-manager --add-repo https://dl.rockylinux.org/vault/rocky/8.5/AppStream/\$basearch/os/ \
&& rpm --import https://dl.rockylinux.org/pub/rocky/RPM-GPG-KEY-Rocky-8 \
&& dnf install -y yum-utils ccache gcc-toolset-10-gcc-10.2.1-8.2.el8 gcc-toolset-10-gcc-c++-10.2.1-8.2.el8 gcc-toolset-10-binutils-2.35-11.el8 \
&& dnf install -y ccache \
&& yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/x86_64/cuda-rhel8.repo
ENV PATH=/opt/rh/gcc-toolset-10/root/usr/bin:$PATH
ARG VULKANVERSION
RUN wget https://sdk.lunarg.com/sdk/download/${VULKANVERSION}/linux/vulkansdk-linux-x86_64-${VULKANVERSION}.tar.xz -O /tmp/vulkansdk-linux-x86_64-${VULKANVERSION}.tar.xz \
&& tar xvf /tmp/vulkansdk-linux-x86_64-${VULKANVERSION}.tar.xz \
&& dnf -y install ninja-build \
&& ln -s /usr/bin/python3 /usr/bin/python \
&& /${VULKANVERSION}/vulkansdk -j 8 vulkan-headers \
&& /${VULKANVERSION}/vulkansdk -j 8 shaderc
RUN cp -r /${VULKANVERSION}/x86_64/include/* /usr/local/include/ \
&& cp -r /${VULKANVERSION}/x86_64/lib/* /usr/local/lib
ENV PATH=/${VULKANVERSION}/x86_64/bin:$PATH
FROM --platform=linux/arm64 almalinux:8 AS base-arm64
# install epel-release for ccache
RUN yum install -y yum-utils epel-release \
&& dnf install -y clang ccache \
&& yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel8/sbsa/cuda-rhel8.repo
ENV CC=clang CXX=clang++
FROM base-${TARGETARCH} AS base
ARG CMAKEVERSION
RUN curl -fsSL https://github.com/Kitware/CMake/releases/download/v${CMAKEVERSION}/cmake-${CMAKEVERSION}-linux-$(uname -m).tar.gz | tar xz -C /usr/local --strip-components 1
COPY CMakeLists.txt CMakePresets.json .
COPY ml/backend/ggml/ggml ml/backend/ggml/ggml
ENV LDFLAGS=-s
FROM base AS cpu
RUN dnf install -y gcc-toolset-11-gcc gcc-toolset-11-gcc-c++
ENV PATH=/opt/rh/gcc-toolset-11/root/usr/bin:$PATH
RUN --mount=type=cache,target=/root/.ccache \
cmake --preset 'CPU' \
&& cmake --build --parallel --preset 'CPU' \
&& cmake --install build --component CPU --strip --parallel 8
FROM base AS cuda-12
ARG CUDA12VERSION=12.8
RUN dnf install -y cuda-toolkit-${CUDA12VERSION//./-}
ENV PATH=/usr/local/cuda-12/bin:$PATH
RUN --mount=type=cache,target=/root/.ccache \
cmake --preset 'CUDA 12' \
&& cmake --build --parallel --preset 'CUDA 12' \
&& cmake --install build --component CUDA --strip --parallel 8
FROM base AS rocm-6
ENV PATH=/opt/rocm/hcc/bin:/opt/rocm/hip/bin:/opt/rocm/bin:/opt/rocm/hcc/bin:$PATH
RUN --mount=type=cache,target=/root/.ccache \
cmake --preset 'ROCm 6' \
&& cmake --build --parallel --preset 'ROCm 6' \
&& cmake --install build --component HIP --strip --parallel 8
FROM --platform=linux/arm64 nvcr.io/nvidia/l4t-jetpack:${JETPACK5VERSION} AS jetpack-5
ARG CMAKEVERSION
RUN apt-get update && apt-get install -y curl ccache \
&& curl -fsSL https://github.com/Kitware/CMake/releases/download/v${CMAKEVERSION}/cmake-${CMAKEVERSION}-linux-$(uname -m).tar.gz | tar xz -C /usr/local --strip-components 1
COPY CMakeLists.txt CMakePresets.json .
COPY ml/backend/ggml/ggml ml/backend/ggml/ggml
RUN --mount=type=cache,target=/root/.ccache \
cmake --preset 'JetPack 5' \
&& cmake --build --parallel --preset 'JetPack 5' \
&& cmake --install build --component CUDA --strip --parallel 8
FROM --platform=linux/arm64 nvcr.io/nvidia/l4t-jetpack:${JETPACK6VERSION} AS jetpack-6
ARG CMAKEVERSION
RUN apt-get update && apt-get install -y curl ccache \
&& curl -fsSL https://github.com/Kitware/CMake/releases/download/v${CMAKEVERSION}/cmake-${CMAKEVERSION}-linux-$(uname -m).tar.gz | tar xz -C /usr/local --strip-components 1
COPY CMakeLists.txt CMakePresets.json .
COPY ml/backend/ggml/ggml ml/backend/ggml/ggml
RUN --mount=type=cache,target=/root/.ccache \
cmake --preset 'JetPack 6' \
&& cmake --build --parallel --preset 'JetPack 6' \
&& cmake --install build --component CUDA --strip --parallel 8
FROM base AS vulkan
RUN --mount=type=cache,target=/root/.ccache \
cmake --preset 'Vulkan' \
&& cmake --build --parallel --preset 'Vulkan' \
&& cmake --install build --component Vulkan --strip --parallel 8
FROM base AS build
WORKDIR /go/src/github.com/ollama/ollama
COPY go.mod go.sum .
RUN curl -fsSL https://golang.org/dl/go$(awk '/^go/ { print $2 }' go.mod).linux-$(case $(uname -m) in x86_64) echo amd64 ;; aarch64) echo arm64 ;; esac).tar.gz | tar xz -C /usr/local
ENV PATH=/usr/local/go/bin:$PATH
RUN go mod download
COPY . .
ARG GOFLAGS="'-ldflags=-w -s'"
ENV CGO_ENABLED=1
RUN --mount=type=cache,target=/root/.cache/go-build \
go build -trimpath -buildmode=pie -o /bin/ollama .
FROM --platform=linux/amd64 scratch AS amd64
COPY --from=cuda-12 dist/lib/ollama /lib/ollama
COPY --from=vulkan dist/lib/ollama/vulkan /lib/ollama/vulkan
FROM --platform=linux/arm64 scratch AS arm64
COPY --from=cuda-12 dist/lib/ollama /lib/ollama/cuda_sbsa
COPY --from=jetpack-5 dist/lib/ollama /lib/ollama/cuda_jetpack5
COPY --from=jetpack-6 dist/lib/ollama /lib/ollama/cuda_jetpack6
FROM scratch AS rocm
COPY --from=rocm-6 dist/lib/ollama /lib/ollama
FROM ${FLAVOR} AS archive
ARG VULKANVERSION
COPY --from=cpu dist/lib/ollama /lib/ollama
COPY --from=build /bin/ollama /bin/ollama
FROM ubuntu:24.04
RUN apt-get update \
&& apt-get install -y ca-certificates libvulkan1 \
&& apt-get clean \
&& rm -rf /var/lib/apt/lists/*
COPY --from=archive /bin /usr/bin
ENV PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
COPY --from=archive /lib/ollama /usr/lib/ollama
ENV LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64
ENV NVIDIA_DRIVER_CAPABILITIES=compute,utility
ENV NVIDIA_VISIBLE_DEVICES=all
ENV OLLAMA_HOST=0.0.0.0:11434
EXPOSE 11434
ENTRYPOINT ["/bin/ollama"]
CMD ["serve"]
Reference in New Issue View Git Blame Copy Permalink