Grandline AI

Pull to refresh

Settings

Model System Prompt Temperature: 0.20 Max Tokens Stream responses Tech mode (raw request/response, latency, tokens) Voice (mic dictation + read replies aloud) Uses your browser's built-in speech engine ÃƒÂ¢Ã¢â€šÂ¬Ã¢â‚¬Â free, on-device. Works best in Chrome (Android/desktop).

👤 User access

Mode Direct: your API key is used from this phone.

Base URL Your own OpenAI-compatible endpoint. The app calls {baseUrl}/chat/completions. API Key

Stored only on this device.

🔑 Admin access

○ Admin mode OFF

Manage who can use your proxy. The admin token only lives on the main phone. Clearing it logs this device out of admin mode.

Proxy URL Admin Token

Set once on the proxy (env var ADMIN_TOKEN). Only your main phone keeps it.

Add, edit, or delete models in the picker. Per-model defaults override General settings when that model is selected.

Log in with your email to sync your chats across all your phones. Your chats are encrypted in transit and synced through your proxy.

Theme Accent Mild green by default.

Wallpaper

Image is stored only on this device (IndexedDB). Pick any image you have rights to use.

Wallpaper Dim: 55%

Your chats are saved on this device only. Back them up to a file on your phone, and restore them on any phone.