Gemini’s Computer Use tool is now generally available across the Interactions and GenerateContent APIs. It turns screenshots into clicks and keystrokes for browser, mobile and desktop agents; developers still supply the execution environment.

Google’s Computer Use overview diagram: send a screenshot and context to the model, receive its response, execute the action in a client app and capture the new screen.

Published 30 Oct 2026 · 13:35 UTC

Google Gemini API release notes ↗ · Source dated 30 Oct 2026

Gemini’s Computer Use tool is now generally available across the Interactions and GenerateContent APIs. It turns screenshots into clicks and keystrokes for browser, mobile and desktop agents; developers still supply the execution environment. · ournews.ai