vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

remove flags

Browse source
This commit is contained in:
Robert Washbourne 2023-11-23 23:58:40 -05:00
parent 22804439d2
commit 0e2c422b11
1 changed files with 4 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

8
.devops/main-cuda.Dockerfile
View file

@ -23,9 +23,9 @@ COPY . .
ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH} ENV CUDA_DOCKER_ARCH=${CUDA_DOCKER_ARCH}
# Enable cuBLAS # Enable cuBLAS
ENV LLAMA_CUBLAS=1 ENV LLAMA_CUBLAS=1
ENV LLAMA_CUDA_MMV_Y=2 # ENV LLAMA_CUDA_MMV_Y=2
ENV LLAMA_CUDA_DMMV_X=64 # ENV LLAMA_CUDA_DMMV_X=64
ENV LLAMA_CUDA_F16=true # ENV LLAMA_CUDA_F16=true
RUN make RUN make
@ -41,7 +41,7 @@ RUN apt-get install python3 python3-pip
RUN pip install --install-option="--prefix=/install" runpod RUN pip install --install-option="--prefix=/install" runpod
FROM ${BASE_CUDA_RUN_CONTAINER} as runtime FROM ${BASE_CUDA_RUN_CONTAINER} as runtime
COPY --from=build /install /usr/local
COPY --from=build /app/server /server COPY --from=build /app/server /server
COPY --from=build /model.gguf model.gguf COPY --from=build /model.gguf model.gguf
COPY --from=build /app/models models COPY --from=build /app/models models