demo-llm / Dockerfile
Petro
First model version
2d3c757
raw
history blame
357 Bytes
FROM python:3.9
COPY requirements.txt ./requirements.txt
RUN python -m pip install -U pip && \
python -m pip install -r requirements.txt && \
python -m pip cache purge
COPY ./main.py /code/main.py
COPY ./zephyr-7b-beta.Q4_K_S.gguf /code/zephyr-7b-beta.Q4_K_S.gguf
WORKDIR /code
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]