Google DeepMind opens Genie 4 to AI Ultra users
Understand this story
Google DeepMind is rolling out a new Project Genie experience to eligible Google AI Ultra subscribers, opening the Genie 4 world model for the first time. Google AI Pro members can join a waitlist on the same page. According to The Decoder, Genie 4 was previously available only as a research preview and is now open to AI Ultra subscribers; it supports real-time changes to weather, characters or plot via text and can generate real-time sound effects that change with the scene. The update also includes upgrades to Genie 3, a Project Genie agent that helps users build scenes, and a community gallery for discovering, exploring and remixing worlds. Genie 3 remains in the Project Genie model selector for those who need longer runs or panoramic seed spaces.
AI overview based on source reports · Updated 2 days ago
Coverage timeline
Follow the coverage to explore different perspectives.
- The DecoderGoogle Deepmind's Project Genie update brings Genie 4, chainable worlds, and a community gallery for remixing
Google DeepMind is releasing Genie 4, its most advanced world model for generating interactive 3D worlds, now open to Google AI Ultra subscribers after previously being a research preview. Genie 4 lets users reshape worlds through real-time text input, such as changing weather, characters, or storylines, and generates native real-time sound that adapts to the scene. The older Genie 3 gains longer sessions of up to three minutes, 360-degree panorama support, and chainable worlds, while a new AI agent builds worlds from spoken descriptions and a community gallery lets users share and remix creations.
- Google DeepMindFeaturedGenie 4 is now available
Google is rolling out a new Project Genie experience to eligible Google AI Ultra subscribers, featuring first-time access to Genie 4, upgrades to Genie 3, the Project Genie agent, and a community gallery. Genie 4 adds mid-session prompting to edit environments in real time and pairs visuals with real-time native audio, while Genie 3 gains sessions up to 3 minutes, world chaining, 360-degree panorama grounding, and real-time in-world events. All generated video and audio includes SynthID watermarking, and Google notes Genie 3 and Genie 4 remain frontier research world models with open challenges including camera latency and long-term scene permanence.
Activity for this story
- Current comparable activity
- 6
- Peak comparable activity
- 19Nov 19 04:00
- Change in the past 24 hours
- -50%
Trends compare the same fully observed participants, which may cover fewer participants than current activity. Move the pointer or tap the chart to inspect hourly activity; use the left and right arrow keys with a keyboard.