Dockerfile

FROM python:3.10-slim

ENV PYTHONDONTWRITEBYTECODE=1 \
    PYTHONUNBUFFERED=1 \
    PIP_NO_CACHE_DIR=1 \
    HF_HOME=/data/.huggingface \
    TRANSFORMERS_CACHE=/data/.cache/huggingface \
    HF_HUB_ENABLE_HF_TRANSFER=1 \
    TOKENIZERS_PARALLELISM=false

RUN apt-get update && apt-get install -y --no-install-recommends \
    gcc g++ make cmake git \
    libopenblas-dev \
    && rm -rf /var/lib/apt/lists/*

WORKDIR /app

# Pre-built llama-cpp-python wheel with OpenBLAS
RUN pip install --no-cache-dir \
    https://huggingface.co/Luigi/llama-cpp-python-wheels-hf-spaces-free-cpu/resolve/main/llama_cpp_python-0.3.22-cp310-cp310-linux_x86_64.whl

COPY requirements.txt .
RUN pip install --no-cache-dir -r requirements.txt

COPY . .

EXPOSE 7860
CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]