Google Launches Gemini 2.5 Computer Use Model for Faster UI Control
Google has launched the Gemini 2.5 Computer Use model, now available to developers via the Gemini API. This model, built on Gemini 2.5 Pro's visual understanding and reasoning, excels in web and mobile UI control tasks with lower latency than leading alternatives.
The model, currently in public preview, can be accessed through Facebook AI Studio and Vertex AI. It's primarily optimized for web browsers and mobile UI control, though desktop OS-level control is not yet supported.
Google has equipped the model with safety features and developer controls to minimize risks. It enables agents to interact with user interfaces for tasks like form filling and web navigation. The core capabilities are exposed through the new tool in the Gemini API.
The Gemini 2.5 Computer Use model outperforms competitors on multiple web and mobile control benchmarks. Google's release of this model via the Gemini API allows developers to explore its potential in UI interaction tasks. However, desktop OS-level control is not yet optimized.
