🔍 Read the full analysis: Inside Google’s Genie 4: Creating Worlds You Can Explore on ThorstenMeyerAI.com
Get hardware and tech essentials delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
TL;DR
Google DeepMind presented Genie 4 on October 27, 2026, describing a world model that creates interactive scenes with generated audio and text-directed changes. The system is not generally available; Google says access is planned for trusted testers and that scenes can drift during sessions lasting a few minutes.
Google DeepMind demonstrated Genie 4 on October 27, describing it as a real-time world model that lets users move through generated scenes, alter them with text and hear sound produced alongside the visuals. The system is not yet generally available: DeepMind says it plans to give access to trusted testers, while acknowledging that scenes can drift and that interaction lasts only a few minutes.
In DeepMind’s demonstrations, a user can steer a scene with keyboard or mouse controls and enter instructions to change elements such as weather, camera movement or the appearance of an object or person. The model generates the next moments in response to the scene and the user’s inputs. Unlike a conventional game engine, Genie 4 does not use a stored 3D map of the environment, according to the source account; it predicts what the scene should look like as it continues.
DeepMind says Genie 4 can produce ambient sound, music and dialogue in sync with its generated visuals. It also says the model can represent physical events, including collisions and aerodynamics, and that Gemini 3.8 Flash can help interpret goals, divide them into steps and monitor the resulting frames. These descriptions come from Google’s account and selected demonstrations; DeepMind has not published benchmark figures in the supplied material to quantify the claims.
Access is expected to start with trusted testers. DeepMind says Google AI Ultra subscribers will be able to use the model through Project Genie and that early partners will get access through the Gemini API. The source gives no launch date or API pricing. It reports that DeepMind places continuous interaction at “a few minutes.” TechCrunch and Bloomberg, as cited by heise, describe demonstrations lasting up to about two minutes, but that specific duration is not stated in DeepMind’s own description.
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
From Generated Clips to Interactive Scenes
Genie 4 points toward a different way to create digital environments: rather than rendering a fixed clip or building every part of a game world by hand, users can prompt a scene and then explore and change it while it runs. If the approach becomes reliable and affordable, it could help studios sketch ideas for game prototypes and film previsualization. DeepMind reports those uses as feedback from testers, not as independently verified production results.
The demonstrations also have research implications. DeepMind showed a robotics simulation involving a humanoid performing a checkers task, but the supplied account says no physical robot was involved. That makes the clip evidence of a simulated experiment, not proof that Genie 4 can train or control a robot in the real world. Separately, the growing realism of generated image-and-audio scenes increases the importance of identifying synthetic media. Google says Genie 4 outputs carry an invisible SynthID watermark; the material does not establish how consistently that watermark can be detected after editing or reposting.
Products Worth Considering
Logitech MX Mechanical Full-Size Wireless Mechanical Keyboard, Tactile Quiet Switches, Backlit Keys, Bluetooth, USB-C, Multi-Device, Graphit
LG 34GX90SA-W 34-inch Ultragear WQHD OLED Curved Gaming Monitor
LG UltraGear 45GX900A-B 45-Inch Curved OLED Gaming Monitor
As an affiliate, we earn on qualifying purchases.
How Genie Models Have Developed
Genie 4 follows a series of systems that DeepMind has presented as steps toward interactive generated worlds. The first Genie, introduced in February 2024, made simple 2D environments. Genie 2, announced in December 2024, generated 3D worlds from a single image and maintained them for up to a minute, though real-time use required a reduced-quality version. These milestones are described in the supplied source material.
In August 2025, DeepMind introduced Genie 3 as a real-time model, reporting output at 720p and 24 frames per second and consistency for a few minutes. Google later opened Genie 3 through Project Genie to paying U.S. subscribers, according to the source account. Waymo, Google’s self-driving-car company, already uses the predecessor Genie 3 in its driving simulator, the account says. Those examples provide context for the research direction, but they do not establish that Genie 4 is ready for commercial deployment.
“Details and objects “can shift or drift over time.””
— Google DeepMind, describing Genie 4’s limits
Access, Reliability and Performance
Several practical details remain unannounced. Google has not provided a public release date, API price or broad-access schedule in the supplied material. The account also says DeepMind published no benchmark numbers, leaving open how Genie 4 performs against other tools on visual consistency, responsiveness, audio quality or compute use.
It is also unclear how often the reported drift occurs, how long a scene remains coherent under different kinds of interaction, or whether users can reliably return to the same locations and objects. The demonstrations were selected by DeepMind, and the source notes that the clips were picked and shortened. As a result, they do not establish typical performance across longer or less controlled sessions. Reported use cases such as film and game prototyping are tester feedback; the robotics example is a simulation rather than a real-world robot test.
Google’s Tester Rollout
The next public milestone is access for trusted testers through Project Genie for Google AI Ultra subscribers and through the Gemini API for early partners, as described by DeepMind. No dates or API prices have been provided, so the scope and timing of that rollout remain uncertain. Tester access should provide more evidence about session limits, scene stability and whether generated audio and text-directed edits work beyond the selected announcement clips.
Until more details and independent testing are available, Genie 4 is best understood as a research and demonstration system with potential applications, not as a confirmed replacement for game engines, film production tools or physical robotics testing.
Key Questions
What is Google Genie 4?
Genie 4 is a world model that predicts and generates the next moments of a scene in response to user controls and text instructions. DeepMind says it can produce visual scenes with audio, rather than rendering only a fixed clip.
Can the public use Genie 4 now?
No public release is described in the supplied material. DeepMind says access is planned for trusted testers, including Google AI Ultra subscribers through Project Genie and early Gemini API partners. It has not given a date.
How long can users interact with a Genie 4 world?
DeepMind describes continuous interaction lasting a few minutes. TechCrunch and Bloomberg, as cited by heise, reported scenes of up to about two minutes, but that specific figure is not in DeepMind’s own description.
Can Genie 4 train real robots?
The supplied account describes a robotics demonstration as a simulation; no physical robot was involved. The demonstration does not establish that Genie 4 can train or control robots in real-world conditions.
What limitations has DeepMind disclosed?
DeepMind says that details and objects can shift or drift over time and that sessions last only a few minutes. It has not published benchmark figures in the supplied material to measure consistency or performance.
Source: ThorstenMeyerAI.com
