Gemini's Computer Use tool leaves preview, reaches GA
Google moved its browser, mobile and desktop control tool for Gemini out of preview and into general availability, across the Interactions and GenerateContent APIs.
Google released the Computer Use tool to general availability on 30 October, according to the Gemini API changelog, across the Interactions and GenerateContent APIs. The GA entry itself is one line: no model name, price or new limit attached.
The tool first shipped as a public preview in June, built for Gemini 3.5 Flash, with simplified actions via intents, built-in support for browser, mobile and desktop environments, configurable safety policies and prompt injection detection. Before that, a January preview had added support for gemini-3-pro-preview and gemini-3-flash-preview. None of that detail is restated in the GA note itself, so it’s not guaranteed to carry over unchanged.
Why it matters: if you’ve been building a browser or desktop-controlling agent against the preview version of this tool, you can now point production traffic at it without a preview label attached. That usually comes with firmer support and stability expectations than a preview endpoint gets, though Google doesn’t spell out what changed between preview and GA here.
The caveat: the changelog gives no model name, pricing or limits for the GA release. If any of that governs a decision you’re making, check the Computer Use docs and your account dashboard directly rather than assuming the preview numbers still apply.