AI News ai-agents inference
AI News Info

Genie 4 arrives in Project Genie with mid-session prompting

Project Genie now gives first-time access to Genie 4, which adds mid-session prompting and real-time native audio, while Genie 3 worlds gain longer sessions and panorama seeding.

KT Kusal Tharindu

The new Project Genie experience includes first-time access to Genie 4. Google describes Genie 4 as its most realistic and interactive world model yet, and the release adds mid-session prompting: type a new prompt while navigating a world and the environment is edited in real time. Genie 4 also pairs its visuals with real-time native audio that adapts as the scene unfolds, from ambient soundscapes to character dialogue.

Genie 3 receives upgrades in the same release, with extended sessions that allow exploration for up to 3 minutes per generation, up from 60 seconds. New worlds can be seeded from 360-degree panorama images, and in-world events can be changed with natural-language prompts while you explore.

Why it matters

The Project Genie agent orchestrates Google’s multimodal models to craft initial visuals and custom background scores from a natural-language description. Curation comes with the same release: a new Project Genie gallery brings together curated worlds alongside creative work from the community. Generated video and audio in Project Genie carry SynthID watermarking.

What to do

Eligible Google AI Ultra subscribers can try the new Project Genie experience starting today, and Google AI Pro members can join the waitlist on the same page.

Sources

  1. Genie 4 is now available
3 posts

Discussion

Spotted an error? Send corrections to [email protected] .

Search Oxelan Signal

Type to search the archive.

Search by title, tag, summary or body text.