🔍 Read the full analysis: Google’s Genie 4 Explained: What It Does And Who Can Use It on ThorstenMeyerAI.com
Prime made for students and young adults
- Fast, free delivery for dorm and study essentials
- Prime Video and Amazon Music included
- Member-only deals
TL;DR
Google DeepMind presented Genie 4 on October 27, 2026, describing it as a real-time AI model that generates interactive scenes from user input. It is not generally available: trusted testers can use it now, while access for Google AI Ultra subscribers and Gemini API partners is planned without a date. The demos are company-selected, and DeepMind says scenes can drift and support only a few minutes of continuous interaction.
Google DeepMind presented Genie 4 on October 27, describing an AI model that generates interactive scenes as users move through them and issue text instructions. The model is not yet generally available: DeepMind says it is being tested by trusted users, with later access planned for Google AI Ultra subscribers through Project Genie and selected Gemini API partners.
Unlike a video generator that produces a clip for later viewing, Genie 4 is designed to generate the next moment as a scene is played. Users can move through a generated environment with keyboard and mouse controls, then type instructions to change elements such as weather, objects or camera movement. DeepMind says its Gemini model helps interpret those instructions and monitor the scene.
The company says Genie 4 adds more realistic visuals and physics, as well as audio generated alongside the images, including environmental sounds, music and dialogue. DeepMind also describes an agent-based feature in which Gemini breaks a goal into steps and checks the resulting frames. These are vendor descriptions and demonstrations; DeepMind has not published benchmark figures in the supplied announcement material.
Availability remains limited. DeepMind says the model is currently with trusted testers. It plans to make it available “soon” to Google AI Ultra subscribers through Project Genie and to early Gemini API partners. The source identifies the subscription as a $200-a-month tier, but gives no launch date, API price or details on how many users will be included.
Infographic 01 · How Genie grew up
From flat sketches to worlds with sound, in under three years.
Each step as Google DeepMind described it. Session limits and consistency are different measures, so read the labels.
- Feb 2024
Genie 1Simple 2D worlds11 billion parameters, learned from unlabelled internet videos - Dec 2024
Genie 23D worlds from a single imageConsistent “for up to a minute”; real time only as a slimmed-down, lower-quality version - Aug 2025
Genie 3First real-time Genie720p at 24 frames per second, consistent “for a few minutes”, memory about one minute back - Jan 2026
Project GenieUS AI Ultra subscribers can play60 seconds per world, partly because of compute costs - May 2026Project Genie goes globalIn the $200 AI Ultra tier, for users 18 and older
- 27 Oct 2026
Genie 4Sound, live steering, more realismResearch preview for trusted testers; Ultra subscribers and API partners “coming soon”
Genie 3 vs. Genie 4
As described by Google DeepMind. Genie 4 column = vendor claims unless marked.
| Feature | Genie 3 (Aug 2025) | Genie 4 (Oct 2026) |
|---|---|---|
| Sound | Genie 3: None | Genie 4: Generated with the picture: ambient noise, music, dialogue |
| Steering | Genie 3: Keys, plus “promptable world events” | Genie 4: Free-form text changes while it runs; Gemini adapts and checks them |
| Look | Genie 3: 720p, 24 fps | Genie 4: “Unprecedented realism”; no resolution or frame rate published |
| How long | Genie 3: “A few minutes”; Project Genie capped at 60 s | Genie 4: “A few minutes” (DeepMind); up to ~2 min consistent scene reported |
| Access | Genie 3: Small preview, then Project Genie | Genie 4: Trusted testers; Ultra and API partners “coming soon” |
Sources: Genie paper, arXiv, 23 Feb 2024; Genie 2 blog, 4 Dec 2024; Genie 3 blog, 5 Aug 2025; TechCrunch, 29 Jan 2026; Google, 19 May 2026; Google DeepMind, Genie 4 blog, 27 Oct 2026; Genie 4 model page (limitations, disclaimer); TechCrunch, 27 Oct 2026. Table wording: Thorsten Meyer.
Powered by Thorsten Meyer AI
A New Tool for Interactive Prototypes
If the capabilities work reliably beyond selected demonstrations, Genie 4 could give game developers and filmmakers a faster way to explore environments and scene ideas. DeepMind says testers have used it for film previsualization and game prototypes, but those examples are feedback reported by the company, not independently verified evidence of production-ready tools.
The technology also matters to robotics research, where simulated environments can help researchers test tasks without putting a physical robot into every scenario. DeepMind’s demonstration of a humanoid playing checkers was a simulation, not a real robot performing the task. Waymo, meanwhile, already uses the earlier Genie 3 in its driving simulator, according to the source material.
Genie 4 also adds another source of realistic-looking synthetic video and sound. DeepMind says its outputs carry an invisible SynthID watermark. That may help identify generated media, but the announcement does not explain how reliably the watermark can be detected after editing or re-encoding.
Popular right now
Bose QuietComfort Wireless Bluetooth Headphones with Active Noise Cancellation, Mic, USB‑C Charging, Deep Bass, 24‑Hour Battery – White Smok
Dyson V8 Cyclone Cordless Vacuum
Vitamix 5200 Professional-Grade Blender, 64 oz. Container, Black
As an affiliate, we earn on qualifying purchases.
How Genie Has Developed
Genie is part of DeepMind’s research into world models: systems that predict how a scene changes in response to time and user actions. Rather than rendering a view from a stored 3D map like a conventional game engine, the model generates successive frames based on what it has already produced and the user’s input. That approach can create flexible scenes, but it can also make details shift as a session continues.
The supplied source traces the series from a 2024 research model that generated simple 2D worlds to Genie 2, which created 3D scenes from a single image. Genie 3, introduced in 2025, ran in real time and maintained scenes for a few minutes, according to DeepMind. Google later offered Genie 3 through Project Genie to paying U.S. subscribers, with TechCrunch reporting a 60-second limit per world. Those earlier versions provide context, but their specifications should not be treated as published Genie 4 benchmarks.
““a few minutes of continuous interaction””
— Google DeepMind
Access and Reliability Still Open
Several practical details are not yet public. DeepMind has not stated when wider access will begin, how many testers or API partners are involved, whether the API will have usage limits, or what it will cost. The phrase “coming soon” does not establish a release date or guarantee that every AI Ultra subscriber will receive access at launch.
Performance is also not independently established by the selected demos. DeepMind says scenes can drift and last only a few minutes of continuous interaction. The source material reports that TechCrunch and Bloomberg, as cited by heise, described scenes of up to about two minutes; that figure is not in DeepMind’s own text. No public benchmark data is provided to compare image quality, responsiveness, sound generation or scene consistency with other systems.
The robotics example is a simulation, and claims about film and game uses are reported tester feedback. It remains unclear how well Genie 4 performs on longer tasks, whether generated worlds can preserve precise layouts, and how SynthID behaves when media is modified.
Watch for Access Details
The next concrete developments would be an access date and eligibility rules for Project Genie and the Gemini API, along with pricing and usage limits. Independent testing after access expands will help clarify whether the demonstrations translate into dependable tools for creative work and research.
Until then, Genie 4 should be understood as a limited-access research and product demonstration, not a generally available game engine or a verified robotics-training system. DeepMind’s own stated constraints—scene drift and a few minutes of interaction—remain central to judging what it can do.
Key Questions
What is Google Genie 4?
Genie 4 is an AI world model that generates interactive scenes over time. Users can move through a scene and provide text instructions that change what happens.
Can the public use Genie 4 now?
No. DeepMind says it is currently available to trusted testers. Access for Google AI Ultra subscribers through Project Genie and early Gemini API partners is planned, but no date has been announced.
Does Genie 4 generate sound as well as images?
DeepMind says it generates audio with the scene, including ambient sound, music and dialogue. The company has not provided benchmark results in the supplied announcement material.
What are Genie 4’s known limitations?
DeepMind says objects and details can shift or drift over time, and that the model supports a few minutes of continuous interaction. The company has not published detailed performance benchmarks.
Is the robot demonstration a real robot using Genie 4?
No. The described humanoid checkers example was a simulation; the source material says no physical robot was involved in that demonstration.
Source: ThorstenMeyerAI.com
