Genie 4 world model reaches developers through the Gemini API
A day after opening Genie 4 to Google AI Ultra subscribers, Google made the world model available to developers through the Gemini API's Interactions API and Google AI Studio. Developers can begin with a written prompt, an image or a mix of the two and generate continuous 720p worlds at 24 frames per second with matching audio, then steer them with keyboard controls and live prompts that change weather, lighting, objects or events. Google suggests uses such as prototyping interactive experiences, visualizing spatial designs and training robots or AI agents. Every output carries an invisible SynthID watermark, and pricing is listed in the API documentation.
Source
Related tool
How we write topics: our editors read the official announcement and other reliable reporting, then write a short neutral summary in their own words. We link to the original sources and do not republish them. Spotted a mistake? Tell us. ← All AI news