Dockerfile

FROM python:3.10-slim

WORKDIR /app

# 1. Install wget (essential for the model)
RUN apt-get update && apt-get install -y wget && rm -rf /var/lib/apt/lists/*

# 2. Install llama-cpp-python FORCING the binary (No compilation allowed)
# We use version 0.2.76 as it is highly stable for MPT models
RUN pip install --no-cache-dir fastapi uvicorn
RUN pip install llama-cpp-python==0.2.76 \
    --extra-index-url https://abetlen.github.io/llama-cpp-python/whl/cpu \
    --prefer-binary

# 3. Download the model silently
RUN wget -q -O model.gguf \
    "https://huggingface.co/maddes8cht/mosaicml-mpt-7b-chat-gguf/resolve/main/mosaicml-mpt-7b-chat-Q2_K.gguf?download=true"

# 4. Copy your script
COPY app.py .

EXPOSE 7860

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