2026-01-06 14:26:42 +00:00
ARG BASE_IMAGE = ubuntu:24.04
2024-04-18 15:19:36 -05:00
ARG GRPC_BASE_IMAGE = ${ BASE_IMAGE }
2024-05-22 23:35:39 +02:00
ARG INTEL_BASE_IMAGE = ${ BASE_IMAGE }
2026-01-06 14:26:42 +00:00
ARG UBUNTU_CODENAME = noble
2023-05-30 15:53:37 +02:00
2025-06-15 14:56:52 +02:00
FROM ${BASE_IMAGE } AS requirements
2023-05-29 23:12:27 +02:00
2024-02-08 20:12:51 +01:00
ENV DEBIAN_FRONTEND = noninteractive
2023-05-29 23:12:27 +02:00
RUN apt-get update && \
2024-04-27 12:48:20 -05:00
apt-get install -y --no-install-recommends \
2025-06-26 18:41:38 +02:00
ca-certificates curl wget espeak-ng libgomp1 \
2026-03-13 20:37:15 +00:00
ffmpeg libopenblas0 libopenblas-dev libopus0 sox && \
2024-04-27 12:48:20 -05:00
apt-get clean && \
2024-05-27 22:07:48 +02:00
rm -rf /var/lib/apt/lists/*
2023-10-16 21:46:29 +02:00
2026-01-07 16:34:23 +01:00
# The requirements-drivers target is for BUILD_TYPE specific items. If you need to install something specific to CUDA, or specific to ROCM, it goes here.
FROM requirements AS requirements-drivers
ARG BUILD_TYPE
ARG CUDA_MAJOR_VERSION = 12
ARG CUDA_MINOR_VERSION = 0
ARG SKIP_DRIVERS = false
ARG TARGETARCH
ARG TARGETVARIANT
ENV BUILD_TYPE = ${ BUILD_TYPE }
ARG UBUNTU_VERSION = 2404
RUN mkdir -p /run/localai
RUN echo "default" > /run/localai/capability
# Vulkan requirements
RUN <<EOT bash
if [ " ${ BUILD_TYPE } " = "vulkan" ] && [ " ${ SKIP_DRIVERS } " = "false" ] ; then
apt-get update && \
apt-get install -y --no-install-recommends \
software-properties-common pciutils wget gpg-agent && \
apt-get install -y libglm-dev cmake libxcb-dri3-0 libxcb-present0 libpciaccess0 \
libpng-dev libxcb-keysyms1-dev libxcb-dri3-dev libx11-dev g++ gcc \
libwayland-dev libxrandr-dev libxcb-randr0-dev libxcb-ewmh-dev \
git python-is-python3 bison libx11-xcb-dev liblz4-dev libzstd-dev \
ocaml-core ninja-build pkg-config libxml2-dev wayland-protocols python3-jsonschema \
2026-01-07 21:49:50 +01:00
clang-format qtbase5-dev qt6-base-dev libxcb-glx0-dev sudo xz-utils mesa-vulkan-drivers
if [ "amd64" = " $TARGETARCH " ] ; then
2026-01-12 07:51:26 +01:00
wget "https://sdk.lunarg.com/sdk/download/1.4.335.0/linux/vulkansdk-linux-x86_64-1.4.335.0.tar.xz" && \
tar -xf vulkansdk-linux-x86_64-1.4.335.0.tar.xz && \
rm vulkansdk-linux-x86_64-1.4.335.0.tar.xz && \
2026-01-07 21:49:50 +01:00
mkdir -p /opt/vulkan-sdk && \
2026-01-12 07:51:26 +01:00
mv 1.4.335.0 /opt/vulkan-sdk/ && \
cd /opt/vulkan-sdk/1.4.335.0 && \
2026-01-07 21:49:50 +01:00
./vulkansdk --no-deps --maxjobs \
vulkan-loader \
vulkan-validationlayers \
vulkan-extensionlayer \
vulkan-tools \
shaderc && \
2026-01-12 07:51:26 +01:00
cp -rfv /opt/vulkan-sdk/1.4.335.0/x86_64/bin/* /usr/bin/ && \
cp -rfv /opt/vulkan-sdk/1.4.335.0/x86_64/lib/* /usr/lib/x86_64-linux-gnu/ && \
cp -rfv /opt/vulkan-sdk/1.4.335.0/x86_64/include/* /usr/include/ && \
cp -rfv /opt/vulkan-sdk/1.4.335.0/x86_64/share/* /usr/share/ && \
2026-01-07 21:49:50 +01:00
rm -rf /opt/vulkan-sdk
fi
if [ "arm64" = " $TARGETARCH " ] ; then
mkdir vulkan && cd vulkan && \
curl -L -o vulkan-sdk.tar.xz https://github.com/mudler/vulkan-sdk-arm/releases/download/1.4.335.0/vulkansdk-ubuntu-24.04-arm-1.4.335.0.tar.xz && \
tar -xvf vulkan-sdk.tar.xz && \
rm vulkan-sdk.tar.xz && \
cd 1.4.335.0 && \
cp -rfv aarch64/bin/* /usr/bin/ && \
cp -rfv aarch64/lib/* /usr/lib/aarch64-linux-gnu/ && \
cp -rfv aarch64/include/* /usr/include/ && \
cp -rfv aarch64/share/* /usr/share/ && \
cd ../.. && \
rm -rf vulkan
fi
2026-01-07 16:34:23 +01:00
ldconfig && \
apt-get clean && \
rm -rf /var/lib/apt/lists/* && \
echo "vulkan" > /run/localai/capability
fi
EOT
# CuBLAS requirements
RUN <<EOT bash
if ( [ " ${ BUILD_TYPE } " = "cublas" ] || [ " ${ BUILD_TYPE } " = "l4t" ] ) && [ " ${ SKIP_DRIVERS } " = "false" ] ; then
apt-get update && \
apt-get install -y --no-install-recommends \
software-properties-common pciutils
if [ "amd64" = " $TARGETARCH " ] ; then
curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu${ UBUNTU_VERSION } /x86_64/cuda-keyring_1.1-1_all.deb
fi
if [ "arm64" = " $TARGETARCH " ] ; then
if [ " ${ CUDA_MAJOR_VERSION } " = "13" ] ; then
curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu${ UBUNTU_VERSION } /sbsa/cuda-keyring_1.1-1_all.deb
else
curl -O https://developer.download.nvidia.com/compute/cuda/repos/ubuntu${ UBUNTU_VERSION } /arm64/cuda-keyring_1.1-1_all.deb
fi
fi
dpkg -i cuda-keyring_1.1-1_all.deb && \
rm -f cuda-keyring_1.1-1_all.deb && \
apt-get update && \
apt-get install -y --no-install-recommends \
cuda-nvcc-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } \
libcufft-dev-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } \
libcurand-dev-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } \
libcublas-dev-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } \
libcusparse-dev-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } \
libcusolver-dev-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION }
2026-01-08 23:04:33 +01:00
if [ " ${ CUDA_MAJOR_VERSION } " = "13" ] && [ "arm64" = " $TARGETARCH " ] ; then
2026-01-07 16:34:23 +01:00
apt-get install -y --no-install-recommends \
libcufile-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } libcudnn9-cuda-${ CUDA_MAJOR_VERSION } cuda-cupti-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION } libnvjitlink-${ CUDA_MAJOR_VERSION } -${ CUDA_MINOR_VERSION }
fi
apt-get clean && \
rm -rf /var/lib/apt/lists/* && \
echo " nvidia-cuda- ${ CUDA_MAJOR_VERSION } " > /run/localai/capability
fi
EOT
RUN <<EOT bash
if [ " ${ BUILD_TYPE } " = "cublas" ] && [ " ${ TARGETARCH } " = "arm64" ] ; then
echo " nvidia-l4t-cuda- ${ CUDA_MAJOR_VERSION } " > /run/localai/capability
fi
EOT
# https://github.com/NVIDIA/Isaac-GR00T/issues/343
RUN <<EOT bash
if [ " ${ BUILD_TYPE } " = "cublas" ] && [ " ${ TARGETARCH } " = "arm64" ] ; then
wget https://developer.download.nvidia.com/compute/cudss/0.6.0/local_installers/cudss-local-tegra-repo-ubuntu${ UBUNTU_VERSION } -0.6.0_0.6.0-1_arm64.deb && \
dpkg -i cudss-local-tegra-repo-ubuntu${ UBUNTU_VERSION } -0.6.0_0.6.0-1_arm64.deb && \
cp /var/cudss-local-tegra-repo-ubuntu${ UBUNTU_VERSION } -0.6.0/cudss-*-keyring.gpg /usr/share/keyrings/ && \
apt-get update && apt-get -y install cudss cudss-cuda-${ CUDA_MAJOR_VERSION } && \
wget https://developer.download.nvidia.com/compute/nvpl/25.5/local_installers/nvpl-local-repo-ubuntu${ UBUNTU_VERSION } -25.5_1.0-1_arm64.deb && \
dpkg -i nvpl-local-repo-ubuntu${ UBUNTU_VERSION } -25.5_1.0-1_arm64.deb && \
cp /var/nvpl-local-repo-ubuntu${ UBUNTU_VERSION } -25.5/nvpl-*-keyring.gpg /usr/share/keyrings/ && \
apt-get update && apt-get install -y nvpl
fi
EOT
# If we are building with clblas support, we need the libraries for the builds
RUN if [ " ${ BUILD_TYPE } " = "clblas" ] && [ " ${ SKIP_DRIVERS } " = "false" ] ; then \
apt-get update && \
apt-get install -y --no-install-recommends \
libclblast-dev && \
apt-get clean && \
rm -rf /var/lib/apt/lists/* \
; fi
RUN if [ " ${ BUILD_TYPE } " = "hipblas" ] && [ " ${ SKIP_DRIVERS } " = "false" ] ; then \
apt-get update && \
apt-get install -y --no-install-recommends \
hipblas-dev \
rocblas-dev && \
apt-get clean && \
rm -rf /var/lib/apt/lists/* && \
echo "amd" > /run/localai/capability && \
# I have no idea why, but the ROCM lib packages don't trigger ldconfig after they install, which results in local-ai and others not being able
# to locate the libraries. We run ldconfig ourselves to work around this packaging deficiency
ldconfig \
; fi
RUN if [ " ${ BUILD_TYPE } " = "hipblas" ] ; then \
ln -s /opt/rocm-**/lib/llvm/lib/libomp.so /usr/lib/libomp.so \
; fi
RUN expr " ${ BUILD_TYPE } " = intel && echo "intel" > /run/localai/capability || echo "not intel"
# Cuda
ENV PATH = /usr/local/cuda/bin:${ PATH }
# HipBLAS requirements
ENV PATH = /opt/rocm/bin:${ PATH }
2025-06-26 18:41:38 +02:00
###################################
###################################
# The requirements-core target is common to all images. It should not be placed in requirements-core unless every single build will use it.
2026-01-07 16:34:23 +01:00
FROM requirements-drivers AS build-requirements
2025-06-26 18:41:38 +02:00
2026-01-06 14:26:42 +00:00
ARG GO_VERSION = 1 .25.4
2026-01-06 02:13:48 +02:00
ARG CMAKE_VERSION = 3 .31.10
2025-06-26 18:41:38 +02:00
ARG CMAKE_FROM_SOURCE = false
ARG TARGETARCH
ARG TARGETVARIANT
RUN apt-get update && \
apt-get install -y --no-install-recommends \
build-essential \
ccache \
ca-certificates espeak-ng \
curl libssl-dev \
git \
git-lfs \
2026-03-13 20:37:15 +00:00
libopus-dev pkg-config \
2025-06-26 18:41:38 +02:00
unzip upx-ucl python3 python-is-python3 && \
apt-get clean && \
rm -rf /var/lib/apt/lists/*
# Install CMake (the version in 22.04 is too old)
RUN <<EOT bash
2025-07-18 08:59:33 -04:00
if [ " ${ CMAKE_FROM_SOURCE } " = "true" ] ; then
2025-06-26 18:41:38 +02:00
curl -L -s https://github.com/Kitware/CMake/releases/download/v${ CMAKE_VERSION } /cmake-${ CMAKE_VERSION } .tar.gz -o cmake.tar.gz && tar xvf cmake.tar.gz && cd cmake-${ CMAKE_VERSION } && ./configure && make && make install
else
apt-get update && \
apt-get install -y \
cmake && \
apt-get clean && \
rm -rf /var/lib/apt/lists/*
fi
EOT
# Install Go
RUN curl -L -s https://go.dev/dl/go${ GO_VERSION } .linux-${ TARGETARCH } .tar.gz | tar -C /usr/local -xz
ENV PATH = $PATH :/root/go/bin:/usr/local/go/bin
2025-07-22 16:31:04 +02:00
# Install grpc compilers
2025-06-26 18:41:38 +02:00
RUN go install google.golang.org/protobuf/cmd/protoc-gen-go@v1.34.2 && \
2025-07-22 16:31:04 +02:00
go install google.golang.org/grpc/cmd/protoc-gen-go-grpc@1958fcbe2ca8bd93af633f11e97d44e567e945af
2025-06-26 18:41:38 +02:00
COPY --chmod= 644 custom-ca-certs/* /usr/local/share/ca-certificates/
RUN update-ca-certificates
RUN test -n " $TARGETARCH " \
|| ( echo 'warn: missing $TARGETARCH, either set this `ARG` manually, or run using `docker buildkit`' )
# Use the variables in subsequent instructions
RUN echo " Target Architecture: $TARGETARCH "
RUN echo " Target Variant: $TARGETVARIANT "
2026-01-07 16:34:23 +01:00
2025-06-26 18:41:38 +02:00
WORKDIR /build
2023-06-26 16:34:03 -04:00
###################################
###################################
2024-05-22 23:35:39 +02:00
# Temporary workaround for Intel's repository to work correctly
# https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/APT-Repository-not-working-signatures-invalid/m-p/1599436/highlight/true#M36143
# This is a temporary workaround until Intel fixes their repository
FROM ${INTEL_BASE_IMAGE } AS intel
2026-01-07 21:55:18 +01:00
ARG UBUNTU_CODENAME = noble
2024-05-22 23:35:39 +02:00
RUN wget -qO - https://repositories.intel.com/gpu/intel-graphics.key | \
gpg --yes --dearmor --output /usr/share/keyrings/intel-graphics.gpg
2026-01-06 14:26:42 +00:00
RUN echo " deb [arch=amd64 signed-by=/usr/share/keyrings/intel-graphics.gpg] https://repositories.intel.com/gpu/ubuntu ${ UBUNTU_CODENAME } /lts/2350 unified " > /etc/apt/sources.list.d/intel-graphics.list
2024-03-29 16:32:40 -05:00
RUN apt-get update && \
2024-04-27 12:48:20 -05:00
apt-get install -y --no-install-recommends \
2025-07-20 22:52:45 +02:00
intel-oneapi-runtime-libs && \
2024-03-29 16:32:40 -05:00
apt-get clean && \
rm -rf /var/lib/apt/lists/*
###################################
###################################
2024-08-14 03:06:41 -04:00
# The builder-base target has the arguments, variables, and copies shared between full builder images and the uncompiled devcontainer
2025-06-26 18:41:38 +02:00
FROM build-requirements AS builder-base
2023-05-30 15:53:37 +02:00
2026-03-19 21:40:51 +01:00
ARG GO_TAGS = "auth"
2023-10-19 13:50:40 +02:00
ARG GRPC_BACKENDS
2024-03-17 09:39:20 -05:00
ARG MAKEFLAGS
2024-08-14 03:06:41 -04:00
ARG LD_FLAGS = "-s -w"
2025-06-26 18:41:38 +02:00
ARG TARGETARCH
ARG TARGETVARIANT
2023-10-19 13:50:40 +02:00
ENV GRPC_BACKENDS = ${ GRPC_BACKENDS }
2023-05-30 15:53:37 +02:00
ENV GO_TAGS = ${ GO_TAGS }
2024-03-17 09:39:20 -05:00
ENV MAKEFLAGS = ${ MAKEFLAGS }
2023-05-30 15:53:37 +02:00
ENV NVIDIA_DRIVER_CAPABILITIES = compute,utility
ENV NVIDIA_REQUIRE_CUDA = " cuda>= ${ CUDA_MAJOR_VERSION } .0 "
ENV NVIDIA_VISIBLE_DEVICES = all
2024-08-14 03:06:41 -04:00
ENV LD_FLAGS = ${ LD_FLAGS }
2023-05-30 15:53:37 +02:00
2024-08-14 03:06:41 -04:00
RUN echo " GO_TAGS: $GO_TAGS " && echo " TARGETARCH: $TARGETARCH "
2023-06-26 16:34:03 -04:00
2024-08-14 03:06:41 -04:00
WORKDIR /build
2024-04-13 02:37:32 -05:00
2023-07-02 11:14:09 +02:00
2025-07-20 22:52:45 +02:00
# We need protoc installed, and the version in 22.04 is too old.
2024-05-28 10:34:59 +02:00
RUN <<EOT bash
if [ "amd64" = " $TARGETARCH " ] ; then
2024-06-10 10:40:02 +02:00
curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v27.1/protoc-27.1-linux-x86_64.zip -o protoc.zip && \
2024-05-28 10:34:59 +02:00
unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
rm protoc.zip
fi
if [ "arm64" = " $TARGETARCH " ] ; then
2024-06-10 10:40:02 +02:00
curl -L -s https://github.com/protocolbuffers/protobuf/releases/download/v27.1/protoc-27.1-linux-aarch_64.zip -o protoc.zip && \
2024-05-28 10:34:59 +02:00
unzip -j -d /usr/local/bin protoc.zip bin/protoc && \
rm protoc.zip
fi
EOT
2024-04-27 12:48:20 -05:00
2024-08-20 06:16:21 -04:00
###################################
###################################
2026-03-05 21:47:12 +01:00
# Build React UI
2026-03-10 07:31:03 +01:00
FROM node:25-slim AS react-ui-builder
2026-03-05 21:47:12 +01:00
WORKDIR /app
COPY core/http/react-ui/package*.json ./
RUN npm install
COPY core/http/react-ui/ ./
RUN npm run build
###################################
###################################
2025-05-25 21:25:05 +01:00
# Compile backends first in a separate stage
FROM builder-base AS builder-backends
2025-06-26 18:41:38 +02:00
ARG TARGETARCH
ARG TARGETVARIANT
2024-08-20 06:16:21 -04:00
2024-04-23 11:43:00 -05:00
WORKDIR /build
2024-07-01 22:50:36 +02:00
2025-05-25 21:25:05 +01:00
COPY ./Makefile .
COPY ./backend ./backend
COPY ./go.mod .
COPY ./go.sum .
COPY ./.git ./.git
# Some of the Go backends use libs from the main src, we could further optimize the caching by building the CPP backends before here
COPY ./pkg/grpc ./pkg/grpc
COPY ./pkg/utils ./pkg/utils
COPY ./pkg/langchain ./pkg/langchain
2024-09-10 02:57:16 -04:00
2025-05-25 21:25:05 +01:00
RUN ls -l ./
2025-07-22 16:31:04 +02:00
RUN make protogen-go
2025-05-25 21:25:05 +01:00
# The builder target compiles LocalAI. This target is not the target that will be uploaded to the registry.
# Adjustments to the build process should likely be made here.
FROM builder-backends AS builder
WORKDIR /build
COPY . .
2024-09-10 02:57:16 -04:00
2026-03-05 21:47:12 +01:00
# Copy pre-built React UI
COPY --from= react-ui-builder /app/dist ./core/http/react-ui/dist
2024-07-01 22:50:36 +02:00
## Build the binary
2025-05-05 17:30:00 +02:00
## If we're on arm64 AND using cublas/hipblas, skip some of the llama-compat backends to save space
## Otherwise just run the normal build
2025-07-18 13:24:12 +02:00
RUN make build
2023-05-30 15:53:37 +02:00
2023-06-26 16:34:03 -04:00
###################################
###################################
2024-08-14 03:06:41 -04:00
# The devcontainer target is not used on CI. It is a target for developers to use locally -
# rather than copying files it mounts them locally and leaves building to the developer
FROM builder-base AS devcontainer
2024-08-20 06:16:21 -04:00
COPY .devcontainer-scripts /.devcontainer-scripts
2024-08-14 03:06:41 -04:00
2024-08-20 06:16:21 -04:00
RUN apt-get update && \
apt-get install -y --no-install-recommends \
2025-06-26 18:41:38 +02:00
ssh less
2024-09-24 03:32:48 -04:00
# For the devcontainer, leave apt functional in case additional devtools are needed at runtime.
2024-08-20 06:16:21 -04:00
2024-08-14 03:06:41 -04:00
RUN go install github.com/go-delve/delve/cmd/dlv@latest
2024-08-20 06:16:21 -04:00
RUN go install github.com/mikefarah/yq/v4@latest
2024-08-14 03:06:41 -04:00
###################################
###################################
2024-04-30 03:12:19 -05:00
# This is the final target. The result of this target will be the image uploaded to the registry.
# If you cannot find a more suitable place for an addition, this layer is a suitable place for it.
2026-01-07 16:34:23 +01:00
FROM requirements-drivers
2023-06-13 01:39:38 -05:00
ENV HEALTHCHECK_ENDPOINT = http://localhost:8080/readyz
2023-06-04 14:00:21 +02:00
2024-06-19 17:50:49 +02:00
ARG CUDA_MAJOR_VERSION = 12
2023-10-19 13:50:40 +02:00
ENV NVIDIA_DRIVER_CAPABILITIES = compute,utility
ENV NVIDIA_REQUIRE_CUDA = " cuda>= ${ CUDA_MAJOR_VERSION } .0 "
ENV NVIDIA_VISIBLE_DEVICES = all
2025-06-26 18:41:38 +02:00
WORKDIR /
2023-06-04 14:00:21 +02:00
2025-06-26 18:41:38 +02:00
COPY ./entrypoint.sh .
2023-11-25 08:48:24 +01:00
2023-10-16 21:46:29 +02:00
# Copy the binary
2023-05-30 15:53:37 +02:00
COPY --from= builder /build/local-ai ./
2026-03-13 20:37:15 +00:00
# Copy the opus shim if it was built
RUN --mount= from = builder,src= /build/,dst= /mnt/build \
if [ -f /mnt/build/libopusshim.so ] ; then cp /mnt/build/libopusshim.so ./; fi
2023-09-04 19:25:23 +02:00
2024-01-09 08:55:43 +01:00
# Make sure the models directory exists
2026-03-14 14:46:52 +00:00
RUN mkdir -p /models /backends /data
2024-01-09 08:55:43 +01:00
2023-05-26 18:34:02 +02:00
# Define the health check command
2023-05-30 12:00:30 +02:00
HEALTHCHECK --interval= 1m --timeout= 10m --retries= 10 \
2024-04-23 11:43:00 -05:00
CMD curl -f ${ HEALTHCHECK_ENDPOINT } || exit 1
2024-05-26 00:56:06 -07:00
2026-03-14 14:46:52 +00:00
VOLUME /models
2023-04-27 18:45:24 +02:00
EXPOSE 8080
2025-06-26 18:41:38 +02:00
ENTRYPOINT [ "/entrypoint.sh" ]
