Dockerfile

# Use a lightweight Python base
FROM python:3.10-slim

# Set working directory
WORKDIR /app

# Install system tools (curl is needed to install Ollama)
RUN apt-get update && apt-get install -y curl && rm -rf /var/lib/apt/lists/*

# Install Ollama
RUN curl -fsSL https://ollama.com/install.sh | sh

# Install Python API requirements
RUN pip install --no-cache-dir fastapi uvicorn requests

# Copy your API code
COPY . .

# Create a start script to handle the background process
RUN echo '#!/bin/bash' > start.sh && \
    echo 'ollama serve &' >> start.sh && \
    echo 'sleep 5' >> start.sh && \
    echo 'ollama pull granite4:latest' >> start.sh && \
    echo 'ollama pull llama3.2:latest' >> start.sh && \
    echo 'ollama pull gemma3:latest' >> start.sh && \
    echo 'uvicorn api:app --host 0.0.0.0 --port 7860' >> start.sh && \
    chmod +x start.sh

# Hugging Face expects port 7860
EXPOSE 7860

# Run the start script
CMD ["./start.sh"]