FROM python:3.11-slim

WORKDIR /app

RUN apt-get update && apt-get install -y --no-install-recommends \
    build-essential cmake && \
    rm -rf /var/lib/apt/lists/*

# Single-threaded build + disable BLAS → ~3GB peak RAM (fits cpu-basic)
ENV CMAKE_BUILD_PARALLEL_LEVEL=1 \
    CMAKE_ARGS="-DGGML_BLAS=OFF -DGGML_NATIVE=OFF" \
    PYTHONUNBUFFERED=1 \
    GRADIO_SERVER_NAME=0.0.0.0 \
    GRADIO_SERVER_PORT=7860

RUN pip install --no-cache-dir llama-cpp-python==0.3.20

RUN pip install --no-cache-dir "gradio>=5.0.0" "huggingface_hub>=0.23.0"

COPY app.py .

EXPOSE 7860

CMD ["python", "app.py"]