diff options
| author | dylan <canardleteer@users.noreply.github.com> | 2023-07-07 11:25:25 -0700 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-07-07 21:25:25 +0300 |
| commit | 84525e7962bee0abef91108948bbf7f7bfdcf421 (patch) | |
| tree | e2e732cb057398249b7b98f46d57d3208a10a2e5 /.devops/main-cuda.Dockerfile | |
| parent | a7e20edf2266169ccd97a4eb949a593d628fbd64 (diff) | |
docker : add support for CUDA in docker (#1461)
Co-authored-by: canardleteer <eris.has.a.dad+github@gmail.com>
Co-authored-by: Georgi Gerganov <ggerganov@gmail.com>
Diffstat (limited to '.devops/main-cuda.Dockerfile')
| -rw-r--r-- | .devops/main-cuda.Dockerfile | 32 |
1 files changed, 32 insertions, 0 deletions
diff --git a/.devops/main-cuda.Dockerfile b/.devops/main-cuda.Dockerfile new file mode 100644 index 0000000..30c0119 --- /dev/null +++ b/.devops/main-cuda.Dockerfile @@ -0,0 +1,32 @@ +ARG UBUNTU_VERSION=22.04 +# This needs to generally match the container host's environment. +ARG CUDA_VERSION=11.7.1 +# Target the CUDA build image +ARG BASE_CUDA_DEV_CONTAINER=nvidia/cuda:${CUDA_VERSION}-devel-ubuntu${UBUNTU_VERSION} +# Target the CUDA runtime image +ARG BASE_CUDA_RUN_CONTAINER=nvidia/cuda:${CUDA_VERSION}-runtime-ubuntu${UBUNTU_VERSION} + +FROM ${BASE_CUDA_DEV_CONTAINER} as build + +# Unless otherwise specified, we make a fat build. +ARG CUDA_DOCKER_ARCH=all + +RUN apt-get update && \ + apt-get install -y build-essential + +WORKDIR /app + +COPY . . + +# Set nvcc architecture +ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH} +# Enable cuBLAS +ENV LLAMA_CUBLAS=1 + +RUN make + +FROM ${BASE_CUDA_RUN_CONTAINER} as runtime + +COPY --from=build /app/main /main + +ENTRYPOINT [ "/main" ] |