Computer use comes to the Mac app
Google announced in an official blog post on Nov. 19 (Nov. 20 Japan time) that it has added "computer use" to the Gemini app for Mac. The feature has the AI click, type, navigate menus and fill in forms on the user's behalf, letting users hand off multi-step tasks that span local files, Mac apps and the web. Google said users only need to describe what they need.
The Gemini app's official account on X also announced it the same day. In an accompanying video, Gemini operates the 3D software "Blender" and turns a quick sketch into a 3D model with a rig, the skeleton used to animate it.
With computer use in the Gemini app for Mac, you can now hand off tasks across your folders, apps, and the web, while your screen and cursor stay yours. Watch Gemini click, scroll, and type in a live background window, turning a quick sketch into a rigged 3D model in Blender.
Operations run in a background window, leaving the screen and cursor to the user
According to Google, Gemini works in a background window isolated from the user's screen. A live preview of its actions appears inside the chat. The user's main screen and cursor stay free, and users can keep typing, designing or taking calls while it works, the company said.
Users can choose which apps Gemini can access. They can watch each step live and pause at any time. Key actions are approved by the user before they are carried out. Google has not specified which operations count as key actions.
Color-grading photos and shipping quotes among the examples
Google gave four examples of how it can be used. Batch color-grading photos in design software, or setting up complex 3D scenes. Extracting weights, dimensions and addresses from local PDFs to fill out multi-box shipping quote forms. Pulling metrics and screenshots from live dashboards across multiple regions and building a presentation in Google Slides. Analyzing clothing photos in a local folder, researching pricing, logging item details and staging listings on resale platforms.
English in the U.S. first, for three paid plans
It is rolling out first in English in the U.S. It requires an Apple Silicon Mac running macOS Sequoia 15.0 or later and a personal Google account for users 18 and older. A subscription to one of the paid plans "Google AI Plus," "Google AI Pro" or "Google AI Ultra" is required.
According to the U.S. pricing page, Plus costs $4.99 a month and Pro costs $19.99 a month. Ultra starts at $99.99 a month, with usage limits 5 times those of Pro. There is also a $199.99-a-month plan with limits 20 times those of Pro.
Google has not disclosed which AI model computer use relies on, limits such as how many times it can be used per day, when it will be offered outside the U.S. or in languages other than English, or whether it supports work accounts. The Gemini app release notes, which run through an entry dated Nov. 17, do not yet mention the feature.
After the API's general availability, the consumer app follows
In release notes dated April 15, Google introduced the Gemini app for Mac. It can be summoned with a keyboard shortcut while using other apps, and users can share a window on screen to get help based on its contents. In an entry dated Oct. 13, Google built its development environment "Google Antigravity" into the Mac app, letting a coding agent build and test code in the background.
For developers, Google made computer use generally available in the Gemini API on Oct. 30. That feature is built into developers' own apps; now ordinary users can hand off the same kind of operations from the Gemini app.
- Gemini app for Mac introducedOfficial
- Google Antigravity built into the Mac appOfficial
- Computer use made generally available in the Gemini APIOfficial
- Computer use added to the Mac appOfficial






