🔍 Read the full analysis: Google Genie 4 Makes Worlds You Can Walk Into. How Does It Hold Up? on ThorstenMeyerAI.com
Get the latest gadgets 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 generates interactive scenes with sound and lets users alter them through text. The demonstrations are company-selected, access is limited to testers, and DeepMind says scenes can drift after a few minutes; it has not published benchmark results or a public release date.
Google DeepMind demonstrated Genie 4 on October 27, presenting an AI model that generates interactive scenes users can move through and change with text prompts. The company says the system can generate visuals and audio together, but it is not publicly available, and DeepMind acknowledges that details can drift during sessions lasting only a few minutes.
DeepMind describes Genie 4 as a world model: rather than rendering a complete clip in advance or drawing from a stored 3D map, it predicts what happens next from the scene so far and the user’s inputs. In the company’s demonstrations, users steer with keyboard or mouse controls and issue instructions such as changing the weather. DeepMind says the model can respond while the scene is running.
The company also says Genie 4 generates sound alongside images, including environmental audio, music and dialogue. It describes Gemini 3.8 Flash as a guiding agent that interprets goals, breaks them into steps and helps monitor the generated scene. These are descriptions from Google’s own announcement; DeepMind has not provided benchmark figures in the source material.
Access is planned first for trusted testers, followed by Google AI Ultra subscribers through Project Genie and early partners using the Gemini API. No launch date or API pricing was given. DeepMind says the model can support “a few minutes of continuous interaction” and that objects and details “can shift or drift over time.” TechCrunch and Bloomberg, as cited by heise, reported demos of up to about two minutes, but that duration is not stated in DeepMind’s own announcement.
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
Interactive Worlds Beyond Game Engines
If the demonstrations translate into dependable tools, Genie 4 could let teams create and explore rough environments without first building a conventional game world or recording a finished video. Film previsualization and game prototyping are among the uses DeepMind says testers have reported, though the source does not provide independent evaluations or details about production use.
The same ability to generate a scene continuously, with corresponding sound, may matter for robotics and simulation research. DeepMind showed a robot-related experiment in which a simulated humanoid played checkers; the report says no physical robot took part. That makes the clip a demonstration of simulation, not evidence that Genie 4 has trained a working robot in the real world.
Generated scenes that look and sound plausible may also make it harder for viewers to identify synthetic media. Google says Genie 4’s video and audio outputs carry invisible SynthID watermarks. The source does not detail how reliably the watermark can be detected or what safeguards apply when generated content is shared outside Google’s tools.
Products Worth Considering
Samsung 27 Odyssey OLED G5 (G50SF) QHD QD-OLED Gaming Monitor
DJI Mic 3 Wireless Microphone System (2 Transmitters, 1 Receiver, Charging Case)
LG UltraGear 27-Inch QHD OLED Gaming Monitor, 240Hz, 0.03ms, G-SYNC Compatible, FreeSync Premium Pro, HDMI 2.1
As an affiliate, we earn on qualifying purchases.
From Genie 3 to Real-Time Audio
Genie 4 follows several generations of work. The first Genie, introduced in February 2024, generated simple 2D worlds from video and was described in its research paper as an 11-billion-parameter model trained on unlabeled internet videos. Genie 2, announced in December 2024, created 3D worlds from a single image and could maintain them for up to a minute, though real-time operation required a reduced-quality version.
In August 2025, DeepMind presented Genie 3 as its first real-time world model, reporting 720p output at 24 frames per second and consistency for a few minutes. Google later opened Genie 3 through Project Genie to paying US subscribers, according to the source. TechCrunch reported a 60-second limit per world, partly attributed to compute costs. Genie 4’s advertised additions are native audio generation and live text-directed changes, alongside claims of more realistic people and physics.
““Details and objects can shift or drift over time.””
— Google DeepMind, describing Genie 4’s limitations
Demo Limits and Access Remain
Genie 4 has not been opened to the public, so outside users cannot yet test whether the scenes respond consistently beyond the selected clips. DeepMind has published no benchmark numbers in the material provided, and there is no independent comparison of visual quality, interaction speed, audio performance or stability.
The precise session limit is also unsettled: DeepMind says “a few minutes,” while TechCrunch and Bloomberg, as cited by heise, report scenes of about two minutes. The source does not establish how the limit varies by prompt, hardware or access tier. It is also unclear what the final product’s availability, API price and content controls will be.
Awaiting Tester Access and Evaluation
Google’s next stated step is to give access to trusted testers, with Google AI Ultra subscribers and early Gemini API partners named as intended groups. DeepMind has not announced when those openings will begin. Wider evaluation will depend on access to the system and on details about session length, costs and reliability that are not yet public.
Until then, the October 27 clips show what DeepMind chose to demonstrate, not how Genie 4 performs across ordinary use. The key test will be whether users can sustain coherent, controllable scenes beyond short demonstrations—and whether the claimed audio and simulation capabilities prove useful outside company-led examples.
Key Questions
What is Google Genie 4?
Genie 4 is a world model that predicts and generates a scene as a user moves through it or gives instructions. DeepMind says it produces visual and audio output in real time.
Can the public use Genie 4 now?
No. DeepMind says access is planned for trusted testers, Google AI Ultra subscribers through Project Genie and early Gemini API partners. It has not announced a public release date.
How long can a Genie 4 scene last?
DeepMind says the model supports a few minutes of continuous interaction and warns that details may drift. TechCrunch and Bloomberg, as cited by heise, reported demos of about two minutes; DeepMind’s own text does not give that specific figure.
Does the robotics demonstration show a real robot?
No. The source describes the checkers example as a simulation involving Apptronik’s Apollo humanoid; it says no physical robot was involved.
Has Google published performance benchmarks for Genie 4?
No benchmark figures are included in the source material. The capabilities described so far are DeepMind’s claims and demonstrations, with independent testing still unavailable.
Source: ThorstenMeyerAI.com
