DeepMind Genie 4 launch
Google DeepMind says it is rolling out a new Project Genie experience, including first-time access to its Genie 4 world model, to eligible Google AI Ultra subscribers starting the day of its announcement.123 Google AI Pro members can join a waitlist.4
Google describes Genie 4 as its most realistic and interactive world model yet, letting users edit a world in real time by typing a new prompt mid-session.5 Google says it is the first time the visuals are paired with real-time native audio, from ambient soundscapes to character dialogue.6
Also included in the update are an agent for Project Genie that assists people in shaping their ideas, plus a shared gallery where users can find, browse and remix creations made by others.7 Genie 3 remains available and was upgraded, with sessions of up to 3 minutes per generation, up from 60 seconds, plus 360-degree panorama grounding.8910
Google says all generated video and audio carries SynthID watermarking and that inputs and outputs are safety-screened, while acknowledging open challenges such as camera latency, complex physical interactions, real-time audio precision and long-term scene permanence.1112
Secondary coverage was brief: Latent Space listed the Genie 4 release among media news alongside Google's Gemini 4 Neon launch, and Zvi Mowshowitz's newsletter only noted that Genie 4 exists.1314
What each sentence rests on
today, we are rolling out a new Project Genie experience to eligible Google AI Ultra subscribers
— Google DeepMind, Genie 4 is now availableThis update unlocks first-time access to Genie 4
— Google DeepMind, Genie 4 is now availableAccess is rolling out to eligible Google AI Ultra subscribers.
— Latent Space, [AINews] Gemini 4 Neon: GDM's Opus/Sol ContenderGoogle AI Pro members can join the waitlist on the same page.
— Google DeepMind, Genie 4 is now availableGenie 4, our most realistic and interactive world model yet, you can now prompt mid-session to change the world you created.
— Google DeepMind, Genie 4 is now availableFor the first time, Genie 4 pairs these visuals with real-time native audio that adapts as the scene unfolds, from ambient soundscapes to character dialogue.
— Google DeepMind, Genie 4 is now availablethe Project Genie agent which helps users build their vision, and a community gallery to discover, explore, and remix worlds
— Google DeepMind, Genie 4 is now availableGenie 3 remains available in Project Genie’s model picker
— Google DeepMind, Genie 4 is now availableExplore for up to 3 minutes per generation (up from 60 seconds)
— Google DeepMind, Genie 4 is now availableSeed new worlds from 360-degree panorama images.
— Google DeepMind, Genie 4 is now available- Quote not reprinted: this page already repeats as much of this article as we allow. — Google DeepMind, Genie 4 is now available
camera latency, intricate physical interactions, real-time audio precision, and long-term scene permanence
— Google DeepMind, Genie 4 is now availableGenie 4 reaches Project Genie. Google’s world model adds native audio and lets users change the world with text mid-session.
— Latent Space, [AINews] Gemini 4 Neon: GDM's Opus/Sol ContenderGoogle Genie 4 exists.
— Zvi Mowshowitz, AI #195: Opening Bell