FROM nvidia/cuda:11.7.1-devel-ubuntu22.04 AS build

RUN apt-get update && apt-get install -y build-essential cmake libboost-dev

ADD ./ggml /build/ggml

RUN mkdir /build/ggml/build

WORKDIR  /build/ggml/build

RUN cmake -DGGML_CUBLAS=ON -DCMAKE_CUDA_COMPILER=/usr/local/cuda/bin/nvcc ..
RUN make codegen-serve

FROM nvidia/cuda:11.7.1-cudnn8-runtime-ubuntu22.04 AS runtime


WORKDIR /app

COPY --from=build /build/ggml/build/bin/codegen-serve /app/codegen-serve

ENV THREADS=4

ENV MODEL="/models/codegen-2B-multi-ggml-4bit-quant.bin"

ENV BATCHSIZE=64

COPY ./run.sh /app/

EXPOSE 18080

CMD /app/run.sh