Genie 4

Released
2026-10-27
Last refreshed
2026-11-18
Status
Researched today
ProprietaryCommercial use: restrictedMultimodal

Genie 4 is a released general LLM work model; evaluate it while provider pricing coverage matures.

Google DeepMind releases · 53 in the last 12 months · this family litChangelog →
Specifications
Family
Genie
Released
2026-10-27
Specialization
video-generation
Openness
Proprietary
License
ProprietaryRestricted
Weights
Not released
Code
Unknown
Created by

Pioneering artificial intelligence research.

London, United Kingdom
Founded 2014
Website
Pricing

No tracked provider pricing is available yet.

About

Genie 4 is Google DeepMind's interactive world model, announced October 27, 2026 as a research preview and made available in Project Genie on November 18, 2026. It generates real-time interactive worlds with native audio, including ambient sound and character dialogue, and supports real-time prompt editing while you navigate. The model card describes a transformer-based system adapted from Gemini Omni Flash, with Gemini used for reasoning and real-time steering. Inputs are text, images, and real-time keyboard and mouse actions; outputs are real-time 720p video at up to 24 frames per second with synchronized audio.

Providers

No active provider routes are tracked for this model yet.

Capabilities

VisionMultimodalAudio

Benchmark peer barsfor Coding

No task-mapped benchmark peers are available for this model yet.

Migration checks

No linked migration route is available for this model yet.