Google bakes computer control directly into Gemini 3.5 Flash, letting the model see and operate your screen

Google has integrated computer-use capabilities into its Gemini 3.5 Flash model, allowing it to interpret screen visuals and perform tasks like navigating software or browsing the web autonomously. By enabling the model to operate computers directly, this update moves beyond simple text generation to functional agent-based automation. Developers can now utilize the Gemini API to build systems that automate complex workflows, positioning the tool to compete with other leading models in software testing and administrative task performance.