Troubleshooting

Common issues and how to debug them.

Server not responding

Confirm the server is running and listening on 11434.

curl http://localhost:11434/api/version

Model is slow

Try a smaller model or a more aggressive quantization. Verify GPU acceleration is enabled if you expect it.

GPU not detected

Ensure the correct drivers are installed and your build supports GPU offload.

Cloud-only tags

If a model tag is cloud-only, it cannot be pulled. Use run/launch instead.

Networking

If running behind a proxy or inside containers, confirm port forwarding and allowed origins.