Dockerfile

FROM python:3.11-slim-bullseye
#FROM nvidia/cuda:12.3.0-devel-ubuntu22.04

ENV HOME=/home/user \
	PATH=/home/user/.local/bin:$PATH \
    TF_ENABLE_ONEDNN_OPTS=0 \
    CMAKE_ARGS="-DLLAMA_BLAS=ON -DLLAMA_BLAS_VENDOR=OpenBLAS" \
    HOST=0.0.0.0 \
    PORT=7860 \
    ORIGINS=*
    
RUN apt-get update && apt-get upgrade -y && apt-get install -y --no-install-recommends \
   python3 \
    python3-pip \
    ninja-build \
    build-essential \
    pkg-config \
    gnupg2 \
    wget

#RUN pip install optimum[onnxruntime]@git+https://github.com/huggingface/optimum.git

WORKDIR /code

RUN chmod 777 .

COPY ./requirements.txt /code/requirements.txt

RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt

#RUN pip install torch==2.1.2 torchvision==0.16.2 torchaudio==2.1.2 --index-url https://download.pytorch.org/whl/cpu \
#    pip install intel-extension-for-pytorch==2.1.100\ 
#    pip install oneccl_bind_pt==2.1.0 --extra-index-url https://pytorch-extension.intel.com/release-whl/stable/cpu/us/

RUN useradd -m -u 1000 user

USER user

WORKDIR $HOME/app

COPY --chown=user . $HOME/app

COPY --chown=user . /home/user/.cache/huggingface/*

RUN chmod 777 .

EXPOSE 7860

CMD ["python", "-m", "main"]