Dockerfile

# Use a lightweight Python runtime
FROM python:3.11-slim

# Set the working directory in the container
WORKDIR /code

# Copy the requirements file and install dependencies
COPY ./requirements.txt /code/requirements.txt
RUN pip install --no-cache-dir --upgrade -r /code/requirements.txt

# Create writable cache directory for Hugging Face models
RUN mkdir -p /code/hf_cache && chmod -R 777 /code/hf_cache

# Set environment variables for Hugging Face cache
ENV HF_HOME=/code/hf_cache

# Copy the application code
COPY ./main.py /code/main.py

# Expose the application port
EXPOSE 7860

# Start the FastAPI application
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "7860", "--timeout-keep-alive", "300"]