{"id":"gemini-computer-use","name":"Gemini computer use","maker":"Google","url":"https://aistudio.google.com/docs/computer-use","repo":"google-gemini/computer-use-preview","summary":"A Gemini API capability in which the model returns interface actions from screenshots so a program can operate a browser.","details":"The developer executes the returned actions with a tool such as Playwright and sends back a new screenshot each turn. A preview repository shows the loop. It requires a Gemini API key.","open_source":false,"hosted":true,"verified":"2026-11-15","category":"browser","category_name":"Browser automation and computer use","page":"https://modelmoot.com/tools/gemini-computer-use","reviewed":"2026-11-15","checked":{"at":"2026-11-15T14:03:44.997Z","address_answers":true,"repository":{"at":"2026-11-15T22:24:09.350Z","archived":false,"licence":"Apache-2.0","language":"Python","last_change":"2026-07-28T14:08:36.000Z","latest_release":null}},"note":"Descriptions are the site's own. Facts under \"checked\" were read automatically from the tool's address and public repository."}