FROM python:3.10-slim ENV HF_HOME=/tmp/hf_cache ENV TRANSFORMERS_CACHE=/tmp/hf_cache ENV LLAMA_CPP_CACHE_DIR=/tmp/hf_cache # Install system deps RUN apt-get update && apt-get install -y git build-essential && rm -rf /var/lib/apt/lists/* WORKDIR /app COPY . /app # Install Python deps RUN pip install --upgrade pip RUN pip install --prefer-binary llama-cpp-python==0.2.90 fastapi uvicorn huggingface-hub transformers # Expose FastAPI port EXPOSE 7860 CMD ["uvicorn", "app:app", "--host", "0.0.0.0", "--port", "7860"]