Settings

Controls how Playto captures and displays text on screen.

• Capture area — choose Cursor Follow (a box that tracks your mouse) or Fixed Region (one or more pinned rectangles), and adjust each region's size and offset.
• Learning mode — switch between standard translation, cloze (fill-in-the-blank practice), or off (capture-only, no overlay).
• Prompt pattern — tell the AI whether captured text should be treated as dialogue, UI labels, or auto-detected (see Translation).
• Overlay appearance — background opacity, font size, alignment, and CEFR difficulty highlighting.
• Auto-save — pick how aggressively Playto saves translated sentences to your word book and journal.

Controls which AI handles translation.

• Mode — Text Reading, Text Reading +, Image Recognition, or Cloud AI (see Translation).
• Local model — pick the model tier for on-device translation; Playto auto-tunes GPU layers based on available VRAM (see Models).
• Cloud AI — enter an API key and choose a provider (Gemini, OpenAI, OpenRouter, or any OpenAI-compatible endpoint); auto vs manual shutter is also configured here. See the Cloud AI setup guide for step-by-step instructions.
• UI language & TTS — change the app's display language and text-to-speech playback speed.

Customize keyboard and mouse bindings for the actions you use most. Defaults are shown in the tab itself and can be overwritten or cleared.

Common bindings: start/stop capture, show/hide overlay, shutter (on-demand capture), cycle cursor preset, nudge and resize the capture region.