Google made the Genie 4 Gemini API, its interactive world model, available to developers through the Gemini API and Google AI Studio on November 19, 2026.
The release comes one day after the model became available to Google AI Ultra subscribers in Project Genie.
What Genie 4 enables
Developers can generate continuous 720p worlds at 24 frames per second with synchronized native audio, starting from a text prompt, a reference image, or both.
Users can navigate with keyboard controls (WASD or arrow keys) and change weather, lighting, objects or events in real time via text prompts. Gemini reasoning keeps the world consistent.
According to Google, the model is useful for prototyping interactive experiences, visualizing spatial concepts, or training AI agents and robotics. All outputs include SynthID watermarking.
Access and pricing
Access requires a paid-tier Gemini API key. The model is available via the Interactions API and in AI Studio.
Reports indicate roughly $0.10 per second of streamed video, with stereo audio included. Sessions can last up to 120 seconds.
Sources: Google official blog, posts from Google DeepMind developers on X.
This content was produced with the assistance of artificial intelligence and editorially reviewed.
Por Grok