Genie 4
Genie 4 is Google DeepMind's world model, announced October 27, 2026. DeepMind describes it as a natively multimodal simulator: visuals and native audio, including music, ambient sound, and dialogue, are generated together.
Provider
Model family
Google Genie
World model
Cost tier
Genie 4
Status
Preview
Release Oct 27, 2026
Why teams choose it
Long-context and Gemini surfaces
Helps when you consolidate analysis in Google-hosted AI paths and rely on large-context ingestion or multimodal prompts.
Long-context analysis
Helps teams summarize, compare, and extract insights from long documents without losing important nuance.
Document-heavy workflows
Useful where teams ingest PDFs, slides, audio, or long threads and need repeatable extraction—not one-off prompting.
Cost-efficient routing
Useful as part of a routing stack where cheap models handle drafts and confirmations and this tier handles genuinely hard passages.
Tradeoffs to know
- Research preview for trusted testers. Not a public Gemini API model.
- No API model ID or price on the October 27 announcement.
Benchmarks
No independent or vendor benchmark scores in this catalog entry.
Technical specs
- Inputs
- text
- Outputs
- video, audio
- License
- Research preview for trusted testers
- Access
- research preview for trusted testers
- Capabilities
- Real-time interactive world simulation
- Native audio generated with the scene
- Text prompts that steer an ongoing simulation
Sourcehttps://deepmind.google/blog/genie-4-a-new-generation-of-world-simulation/
Related models
Gemma 2 27B
Gemma 2 27B is Google’s open-weights Gemma family checkpoint balancing quality and deployability for research and pro…
Gemini 3.1 Pro Preview
Gemini 3.1 Pro Preview is Google's preview Gemini 3-series model for complex multimodal reasoning, software engineeri…
Gemini 2.0 Flash
Gemini 2.0 Flash is Google’s efficiency-oriented multimodal model generation aimed at fast agentic and interactive ex…
Genie 4 FAQ
What is Genie 4?
Genie 4 is Google DeepMind's world model, announced October 27, 2026. DeepMind describes it as a natively multimodal simulator: visuals and native audio, including music, ambient sound, and dialogue, are generated together. Text prompts steer the scene, and the research preview uses keyboard controls for navigation and camera. Gemini Omni anchors appearance and behavior, and other Gemini models supply reasoning for continuity. DeepMind has started a research preview for trusted testers. Consumer and developer release is planned to start with Google AI Ultra subscribers in Project Genie and early-access partners of the Gemini API and Google AI Studio. A Project Genie waitlist is open.
When does Genie 4 fit best?
Interactive world simulation for trusted testers
What should teams watch out for with Genie 4?
Research preview for trusted testers. Not a public Gemini API model.
Explore next
Models, tools, and comparisons that connect to this reference.