changelogGoogle

Google makes Genie 4 available in Gemini API and AI Studio with 720p, 24 fps interactive worlds

TL;DR

Google made Genie 4 available through the Gemini API and Google AI Studio on November 19, 2026. The model generates continuous 720p video at 24 frames per second with synchronized native audio, and worlds respond to keyboard controls and live text prompts.

1 min read
0

Google made Genie 4 available to developers through the Gemini API and Google AI Studio on November 19, 2026. The release gives developers access to interactive worlds that respond to keyboard controls and live text prompts, according to a summary of the announcement published by Unite.AI.

What developers get

  • Output: continuous 720p video at 24 frames per second, with synchronized native audio.
  • Model ID: models/genie-4.
  • Endpoint: the v1beta Interactions API.
  • Access points: the Gemini API and Google AI Studio.
  • Seeding: a world can be started from text, a single reference image, or both.
  • Controls: WASD and arrow keys control movement and camera. Live text prompts can also steer the world as it runs.

What this means

Genie 4 is now reachable through the same Gemini API and Google AI Studio surfaces developers already use for Google models, under the model ID models/genie-4. Because it runs on the v1beta Interactions API, developers building against it are working with a beta endpoint. The combination of keyboard input and live text prompts means a generated world is steered continuously during a session rather than produced as a fixed clip. Native audio is generated in sync with the 720p, 24 fps video stream.

Source: unite.ai ↗

Related Articles

Comments

Loading...