Dockerfile

FROM python:3.11-slim

WORKDIR /app

# Install system dependencies
RUN apt-get update && apt-get install -y \
    ffmpeg \
    && rm -rf /var/lib/apt/lists/*

# Set cache to persistent storage
ENV HF_HOME=/data/.huggingface
ENV HF_HUB_CACHE=/data/.huggingface/hub

# Install your wheel + dependencies
RUN pip install --no-cache-dir \
    https://github.com/Ary5272/llama-cpp-python/releases/download/v0.1.1/llama_cpp_python-0.3.16-cp311-cp311-manylinux2014_x86_64.manylinux_2_17_x86_64.whl \
    gradio \
    faster-whisper \
    huggingface_hub

# Copy app
COPY app.py /app/app.py
COPY entrypoint.sh /entrypoint.sh
RUN chmod +x /entrypoint.sh

EXPOSE 7860

ENTRYPOINT ["/entrypoint.sh"]
CMD ["python", "app.py"]