Hendoyren — Support
Private AI chat for iPhone, iPad, and Mac.
Need help fast? Email hendoyrenapp@gmail.com — we usually reply within a day.
Choosing an AI
Tap the bar under the title to choose who answers. Each choice keeps its own chats.
- Local AI (on-device): uses Apple Intelligence when your device supports it, and a built-in model otherwise. Nothing you type leaves your device. The built-in model downloads once (0.9–1.8GB, depending on your device), then works offline.
- Private Cloud Compute: Apple's server model, for longer conversations and harder questions. Your messages leave your device, and Apple uses them only to answer you and doesn't store them. Hendoyren asks before your first message and never uses it automatically. It needs iOS, iPadOS, or macOS 27 or later on an Apple Intelligence device, and Apple sets a daily limit.
- Your own server: Ollama, LM Studio, or any server with an OpenAI-compatible API. Your messages go directly to the address you enter.
Connecting a server
- Start your server. Ollama uses port
11434and LM Studio uses port1234. To reach it from another device, turn on network access in the server's settings. - Find the address. From iPhone or iPad, use your computer's network address, such as
http://192.168.1.20:11434. Leave off/v1. - Add it. Open Settings, tap Add Server, and enter a name and the address. A token is only needed if your server requires one. If iOS asks about your local network, tap Allow.
- Pick a model above the message box. If your server doesn't list the one you want, edit the server and enter its name.
- Away from home, a local address won't work. Use a private network such as Tailscale with an
https://address.
Troubleshooting
- "Can't reach this server": check that the server is running and accepts connections from other devices. On iPhone or iPad, check Settings → Privacy & Security → Local Network → Hendoyren.
- "Model not found": choose a different model above the message box.
- "This server needs a valid token": enter the token in the sheet that appears. Your message sends again once the server accepts it.
- "Server certificate untrusted": Hendoyren needs a certificate your device already trusts, so self-signed certificates are refused. Use
http://on your local network, or a trusted certificate. - "Server identity changed": the server's TLS certificate rotated. Compare the fingerprints shown and tap Trust New Fingerprint only if you rotated it intentionally.
- Private Cloud Compute isn't listed: it needs iOS, iPadOS, or macOS 27 or later on an Apple Intelligence device.
- Voice not working: add your own ElevenLabs key in Settings → Voice / ElevenLabs.
- Inspect failures: Settings → Diagnostics shows every connection attempt and error. Use Copy Log to share it.