Genie 4 is now available
Summary
Google has launched a new Project Genie experience for eligible Google AI Ultra subscribers, headlined by the debut of Genie 4, described as Google's most realistic and interactive world model yet. It allows users to change their generated worlds mid-session via text prompts in real time, and pairs visuals with real-time native audio that adapts to the unfolding scene, from ambient soundscapes to character dialogue.
The update also brings upgrades to Genie 3, including extended sessions of up to 3 minutes per generation (up from 60 seconds), world chaining, 360-degree panorama image grounding, and real-time in-world events. A new Project Genie agent lets users build interactive universes through natural-language conversation, orchestrating Google's multimodal models to create visuals and custom background scores, while Gemini models offer context-aware suggestions and trigger events during exploration.
A new community gallery offers curated worlds such as "Living Memory," "Dino Field Guide," and "Rock Paper Scissors," plus collaborations like "Villa Giallo" with BLVCKL!GHT and "The Life of Ants" with Anand Gandhi. Users can remix and publish their own creations. Google says all generated video and audio includes SynthID watermarking and undergoes safety screenings, and notes that Genie 3 and Genie 4 remain frontier research models still facing challenges such as camera latency and long-term scene permanence.
(Source:Gemini)