| # 1. Start the Ollama server in the background | |
| ollama serve & | |
| # 2. Record the Process ID (PID) | |
| PID=$! | |
| # 3. Wait for the server to wake up (checks every second) | |
| echo "Waiting for Ollama to start..." | |
| until curl -s http://localhost:7860/api/tags > /dev/null; do | |
| sleep 1 | |
| done | |
| # 4. Pull Gemma 4 (Using the E2B version to ensure it fits in Free Tier RAM) | |
| # Change 'gemma4:e2b' to 'gemma4' if you have a paid GPU Space | |
| echo "Pulling Gemma 4 model..." | |
| ollama pull gemma4:e2b | |
| # 5. Tell the user it's ready | |
| echo "Model pulled! API is ready at port 7860." | |
| # 6. Wait for the background process to finish (keeps the container alive) | |
| wait $PID | |