Dockerfile

FROM python:3.10-slim

WORKDIR /app

# Copy the exact model file into the container
COPY SmolVLM-500M-Instruct-Q8_0.gguf /app/

# Install llama-cpp-python with server
RUN pip install --no-cache-dir llama-cpp-python[server]

EXPOSE 8000

# Run the server with your specific model filename
CMD ["python", "-m", "llama_cpp.server", "--model", "SmolVLM-500M-Instruct-Q8_0.gguf"]