FROM python:3.12

RUN useradd -m -u 1000 app

WORKDIR /home/app

RUN apt update
RUN apt install -y wget make cmake clang git g++
RUN wget https://huggingface.co/neph1/llama-3.1-instruct-bellman-8b-swedish/resolve/main/gguf/llama-3.1-instruct-bellman-8b-swedish-Q8.gguf?download=true -O model.gguf
RUN git clone https://github.com/ggerganov/llama.cpp
RUN mv llama.cpp llama_temp
RUN mv llama_temp/* .
RUN make
RUN apt install socat -y

EXPOSE 7860

CMD ["sh", "-c", "./llama-server -m /home/app/model.gguf -c 8192 --host 0.0.0.0 --port 7860"]