Google DeepMind announced on Tuesday, October 27, 2026, Genie 4, its new world model for real-time interactive simulation. The release brings advances in photorealism, physics and, for the first time in the series, native audio generation — including ambient sounds, music and interactive dialogue.
According to the official announcement, Genie 4 builds on the Gemini family: Gemini Omni anchors realistic appearance and behavior in the simulations, while other models supply the reasoning needed to maintain coherence and continuity. Users can set an event in motion and, in real time, use natural language to steer the scene — changing the weather, adjusting the camera or introducing new characters and objects — while the model simulates how the world responds, sounds and how the people in it move, react and speak.
What changes compared to Genie 3
The predecessor, Genie 3, already allowed explorable worlds from prompts. Genie 4 doubles consistent continuity time (up to about two minutes), improves physics (including aerodynamics and collisions) and human movement, and adds synchronized native audio. Real-time text control and tighter integration with Gemini as a “director” that plans steps and checks frames are other highlights.
The company highlights applications in game prototyping, pre-visualization for film and television, journalism and, especially, safe training environments for robots and agents. In robotics research, Genie 4 was paired with Gemini 4 Argon to simulate scenarios.
Availability and limitations
The model is in research preview for a limited group of trusted testers. It is expected soon for Google AI Ultra subscribers in Project Genie and early-access partners via the Gemini API and AI Studio. There is no exact date or pricing details for developers yet.
DeepMind acknowledges that in longer sessions visual details and object permanence can degrade, and that audio accuracy and text readability still have room for improvement. It is an experimental research model, not a finished product for generating complete games.
Interactive world simulation is seen by the company as an important step toward AGI, enabling the training of systems in varied and safe environments before bringing them into the physical world.
Image credit: Google DeepMind
Sources
- Official announcement on the Google blog
- Detailed post from Google DeepMind
- TechCrunch
- The Verge
- Google DeepMind post on X
Transparency: This content was created, edited or reviewed with the aid of artificial intelligence. Information was cross-checked with public posts on X and sources available on the internet. Consult the original sources to verify the full context.
By GeekikiBot
