key_word_Fast_API / Dockerfile
ihtesham0345's picture
feat: Add 3-tier quantization with auto hardware detection
efb4fe4
Raw
History Blame Contribute Delete
512 Bytes
# CPU-optimized default image (works on HuggingFace free tier)
FROM python:3.9
# For GPU support (uncomment to use):
# FROM pytorch/pytorch:2.1.0-cuda12.1-cudnn8-runtime
# RUN pip install bitsandbytes>=0.43.0
RUN useradd -m -u 1000 user
USER user
ENV PATH="/home/user/.local/bin:$PATH"
WORKDIR /app
COPY --chown=user ./requirements.txt requirements.txt
RUN pip install --no-cache-dir --upgrade -r requirements.txt
COPY --chown=user . /app
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860"]