Genie 4 (GA)

Genie 4, Google DeepMind’s interactive world-generation model, became generally available in the Gemini API on November 19, 2026, a day after it began rolling out in Project Genie.

Lab
Google DeepMind
Release date
Status
Generally available
API model ID
models/genie-4

The Gemini API changelog lists Genie 4 (models/genie-4) as generally available on November 19, 2026, on the Interactions API (v1beta), for real-time interactive world generation: steerable 24 fps, 16:9 video with synchronized 48 kHz stereo audio streamed over WebRTC and WebSockets. Worlds can be seeded from text, a 16:9 reference image or both, and steered live with keyboard controls and mid-session text events. It is available on the paid tier. The model page also accepts the ID genie-4.

Google’s Project Genie post of November 18, 2026, says Genie 4 is rolling out to eligible Google AI Ultra subscribers, with a waitlist for Google AI Pro members. Following our method, the record uses the Gemini API changelog date.

Modalities
Text and 16:9 reference image input; streaming interactive video (24 fps) with 48 kHz stereo audio
Availability
Gemini API paid tier (Interactions API, v1beta). Project Genie for eligible Google AI Ultra subscribers from November 18, 2026; waitlist for Google AI Pro members.
Official source
ai.google.dev
Last checked

Questions

When did Genie 4 become generally available?
In the Gemini API on November 19, 2026; Project Genie began rolling it out to eligible Google AI Ultra subscribers on November 18.
What is the Genie 4 model ID?
models/genie-4 (the API also accepts genie-4).
What does Genie 4 output?
Streaming 16:9 video at 24 fps with synchronized 48 kHz stereo audio.

Source

Release dates, model IDs and limits come from each lab’s own announcement, changelog, documentation or deprecations page, linked on every model with the date it was last checked. How we pick and check sources: editorial standards.

The Model Press

What are you looking for?

Search by headline, topic or keyword.