Connected for iOS

Support

Chat with language models you run yourself — on your local network, over Tailscale, or with Apple Intelligence on supported devices.

Contact

Questions, bug reports, or account issues:

elia.moharer@gmail.com

Quick start

  1. Start a server on your computer (Ollama, LM Studio, oMLX, etc.).
  2. Bind the server to 0.0.0.0 so your phone can reach it on your Wi‑Fi network.
  3. In Connected, tap the brain icon → Find Endpoints and select your server.
  4. Pick a model and start chatting.

On devices with Apple Intelligence, you can chat without a server when no model is selected.

Common issues

My endpoint isn’t showing up
  • Make sure your server is running and has at least one model downloaded.
  • Supported ports: 7590, 11434, 1234, 8000, 8080, 5000.
  • Bind the host to 0.0.0.0 (e.g. OLLAMA_HOST=0.0.0.0 ollama serve).
  • If your server requires an auth key, set it in the app before scanning.
  • Try scanning again once or twice.
The model isn’t responding
  • Confirm the model fits your hardware and is loaded.
  • First responses can take up to a minute while the model loads.
It can’t see my pictures

Some models do not support vision. Check that your model accepts image input before attaching photos.

Restore my purchase

Open the custom endpoint purchase sheet in the app and tap Restore Purchases, or go to Help → Restore Purchases.

Legal