Google DeepMind unveils Genie 4, a steerable real-time world model in research preview
Google DeepMind announced Genie 4 on 27 October 2026, a model that simulates interactive worlds in real time with synchronized visuals and audio. Users can navigate these worlds with keyboard controls, alter them through text prompts, and converse with in-world characters. The system pairs Gemini Omni for world grounding with a Gemini 3.8 Flash agent that plans, checks, and replans multi-step goals. DeepMind acknowledges that visual consistency, object permanence, and long-term coherence remain unresolved, and pricing is unconfirmed. Access is currently limited to trusted testers, with a waitlist for Google AI Ultra subscribers and select API partners.
This is an AI-generated summary. ShortSingh links to the original source for the complete article.
Discussion (0)
Log in to join the discussion and vote.
Log in