Genie 4
Google DeepMind's real-time interactive world model with native audio, prompt steering and two-minute consistent scenes.
Overview
Genie 4 is Google DeepMind's world model for real-time interactive simulation, announced October 27, 2026 as a research preview for trusted testers. Users steer a running world with natural-language prompts, such as changing the weather or camera angle or adding characters, and each world generates audio natively, including ambient sound, music and interactive dialogue. Gemini models plan steps and check frames while Genie renders the visuals; TechCrunch reports scenes stay consistent for up to two minutes, double Genie 3's limit. Google says object drift, real-time audio accuracy and legible in-world text remain open research problems.
This is a sourced directory entry. Features and capabilities are reported in the linked sources; this entry does not document independent Aymorphic testing.
Key features
- Real-time world generation steerable by text prompts mid-simulation
- Native audio generation: ambient sound, music and interactive dialogue
- Gemini acts as planner and frame checker alongside Genie's rendering
- Consistent scenes for up to two minutes, per TechCrunch (Genie 3 was one minute)
- Free-form edits to weather, camera angle, people and objects
Use cases
- Testing AI agents and robot policies in simulated environments before real-world runs
- Prototyping interactive scenes and game-like worlds from text or images
- Research on world models and long-horizon simulation consistency
Pricing
Not priced. Research preview for trusted testers; Google says it will reach Google AI Ultra subscribers in Project Genie and early-access Gemini API / Google AI Studio partners in the coming weeks. Waitlist open via Project Genie.
Pricing reflects the source record and may change. Confirm current terms with the provider.
Dates and sources
- Released
- 27 Oct 2026
- Record updated
- 2 Nov 2026
- Record last checked
- 2 Nov 2026