FROM nvidia/cuda:12.1.1-devel-ubuntu20.04

# Install the deps
ENV DEBIAN_FRONTEND=noninteractive
ENV TZ=Etc/GMT
RUN apt-get update && apt-get install -y --no-install-recommends python3 python3-pip git cmake build-essential
# Get llama-cpp-python
WORKDIR /usr/src

WORKDIR /usr/src/app 

# Build llama-cpp-python w/CuBLAS
RUN CMAKE_ARGS="-DLLAMA_CUBLAS=on" FORCE_CMAKE=1 pip install llama-cpp-python[server]

# We need to set the host to 0.0.0.0 to allow outside access
ENV HOST 0.0.0.0

# Run the server
CMD python3 -m llama_cpp.server