Google DeepMind launched Genie 3, a next-generation world model that creates highly detailed, interactive 3D environments from simple text descriptions.
Genie 3 operates at 720p and 24 fps, supports prompt-based changes within the world, and maintains spatial memory for over a minute, meaning objects stay in place even when users look away and return.
This marks a major leap over its predecessor, which only supported 10–20 seconds of interaction.
Designed for applications such as game development, robotics training, and agent-based simulation, Genie 3 enables users or AI agents to navigate generated worlds in real-time, responding dynamically to environmental changes.
With events, like weather changes or character additions, the model supports improvisation mid-simulation.
DeepMind is releasing Genie 3 as a limited research preview to select academics and creators, positioning it as a key milestone toward artificial general intelligence.
You may also want to check out some of our other recent updates.
Subscribe to Vavoza Insider to access the latest business and marketing insights, news, and trends daily! 🗞️
