Skip to content

Google DeepMind Rolls Out a New Project Genie Experience

Google DeepMindFrom the source·· 1 day agoPicksGenie 4 is now available
Original
Image: Google DeepMind
Who
Google DeepMind
What
rolled out a community gallery and updated Genie 3 in the model picker with sessions up to three minutes
Quote
So today, we are rolling out a new Project Genie experience to eligible Google AI Ultra subscribers, making these worlds far more responsive, creative, and social.

What happened

Google DeepMind 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, a creation agent, and a community gallery. Genie 4 supports real-time world changes from prompts and native audio; Genie 3 sessions can run for up to 3 minutes per generation, up from 60 seconds. Google AI Pro members can join the waitlist.

Why it matters

Eligible Google AI Ultra subscribers can now reshape generated worlds during a session, while Genie 3 supports longer explorations of up to three minutes per generation. The experience also lets users publish and remix worlds in a community gallery.

Reported by 3 sources

    The story so far

    Google DeepMind has made its Genie 4 world model available to eligible Google AI Ultra subscribers within the Project Genie platform. The model, which initially launched as a research preview in October 2026, enables real-time generation of interactive environments that support native audio, interactive dialogue, and environment steering via natural language prompts. Powered by Gemini 3.8 Flash, Genie 4 allows users to alter scenes dynamically—such as adding characters or changing the weather—during active simulation. The broader Project Genie update also introduced a community gallery for sharing creations and extended session durations for the Genie 3 model to three minutes. While Google DeepMind states that Genie 4 improves upon its predecessors in physics and realism, the company notes that maintaining object permanence and physical consistency during extended simulations remains an ongoing research challenge.

    How the story developed · 2

    View full event
    1. 11/19 02:53 (UTC+8)

      Google DeepMind Rolls Out a New Project Genie Experience This report

    2. 10/28 00:56 (UTC+8)
      Google DeepMind announces Genie 4 for real-time world simulation with native audioGoogle DeepMind

    Read next

    See the full storyGoogle DeepMind Releases Genie 4 World Model With Native Audio and Real-Time Interaction
    This story is in 11/19 AI DailyRead the whole issue