Google Opens World Model Genie 4 to Developers: Keyboard Control Latency as Low as 0.2 Seconds, Video Generation at About $6.05/Minute
1 day ago / Read about 0 minute
Author:小编   

On November 19, Google officially made its world model Genie 4 available to developers, who can access and call it through the Gemini API (the Interactions API multi-turn interaction interface) and Google AI Studio. A day earlier, on November 18, Google had already opened Genie 4 to AI Ultra subscribers in its consumer app Project Genie, while AI Pro users can join the waitlist. At the same time, Genie 3 received an upgrade: the maximum duration of a single exploration session has been extended to 3 minutes, and it now supports multi-world chaining and 360-degree panoramic image generation.