Token : Local AI Studio · 更新记录

• New: Lens → Live. LFM3 Omni watches the camera and writes a short comment every few seconds, newest first, until you stop it. Type what to watch for, or just point and look.
• Lens results and Live comments can now be read aloud: turn on "speak in lens" in Voice settings.
• New voices: Supertonic 3 speaks 31 languages, Turkish included, with 10 voices; Paradee is a 9 MB English voice. KittenTTS and Kokoro now say "er" sounds correctly.
• More speech-to-text for hold-to-talk: Apple's on-device model (no download), Parakeet (25 European languages), Nemotron (every app language) and Whisper tiny, small and large-v3-turbo.
• Fixed: the Voices page shows your installed voices again.
• Lens → Clip's button now says Record.

• New app languages: Arabic, Greek, Romanian, Portuguese, and Russian
• Pick assistant reply language separately from the interface language
• Improved on-device tool calling and assistant chat experience
• Faster, more stable MLX and GGUF inference
• Refreshed settings and UI polish

New: Conversation history drawer
Your saved chats now live in a slide-out side panel. Open it with the = button (top-left) or a swipe in from the left edge. Tap a chat to reopen it; swipe a row to delete or export. Tap the dimmed area to close.

New: On-device error analysis
Settings → Privacy → "Analyze errors with AI" feeds the app's local diagnostic report to your loaded model and explains likely causes of any crashes/errors — fully on-device, nothing uploaded.

Polish
Cleaner header, larger tap targets, and better text contrast on the light and dark themes.

Please test
*Open/close the history drawer with both the = button and the left-edge swipe; confirm rows are tappable and the panel is bright (not dimmed)
* Run "Analyze errors with AI" after a model is loaded
* General stability when switching models