Google DeepMind's Genie 4 Adds Real-Time Audio and Two-Minute Worlds

Google DeepMind released Genie 4, its most photorealistic interactive world model to date, in research preview on Tuesday — and this version talks back. The model generates real-time audio matched to the simulated world, including music, ambient noise, and interactive dialogue, and it lets users type prompts mid-simulation to reshape the scene.

The release lands in a consolidating market. Runway shipped its second world model, GWM Worlds 2, a couple of months ago — a playable model that responds dynamically to user input and includes audio generation. Meanwhile, Fei-Fei Li's World Labs is being acquired by AMD, Amazon is in talks to buy Decart, and Rivian just acquired Third Dimension to bring simulation capabilities in-house for self-driving testing. Genie 4 is Google's latest move to stay ahead in that crowd.

What does Genie 4 actually do better?

Plenty, on paper. Genie 3 arrived as a research preview a little over a year ago; its successor doubles the consistent-scene limit to two minutes. Genie 4 also supports freeform edits — changing weather or camera angle in real time, adding new people or objects — with better physics and more human-like characters.

Gemini, Google's flagship AI model, is now more closely integrated, acting as the director that plans steps and checks frames while Genie creates the visuals. In a research experiment tied to Gemini Robotics, a Gemini 4 Argon model broke down tasks into steps that Genie 4 simulated in real time. That's a plausible path to testing robot scenarios in simulation before they run in the physical world.

Tejas Kulkarni, research director at Google, framed the architecture as a hybrid play during a briefing: "What we'll increasingly see in the future are these hybrid world models, where you have a combination of Gemini writing code and text, and Genie ingesting that in real time and simulating reality at different levels of abstraction."

Who is this for?

The pitched commercial use cases are the usual trio: gaming, interactive learning, and robotics. One speculative learning example Google offers is entering a painting of a historical scene and chatting with the painting's subjects about the event portrayed.

But "trusted testers" — the people who get first access to Google's cutting-edge products — are pointing to a different early use case: prototyping films.

"The real time aspect becomes most valuable in the areas where keeping a state of flow in the creativity space becomes incredibly useful," Diego Rivas, a Google DeepMind engineer, said during the press briefing. "For example, in the ideation phase, in the storyboarding phase — anytime you're kind of in need of rapid ideation."

That's a notable signal for practitioners. Google already sells Veo, its AI video generation model that produces high-definition clips with synchronized audio from text, image, or video prompts — a tool that has been growing in popularity among filmmakers. Genie 4 could slot in upstream of Veo as a rapid-prototyping layer: rough interactive worlds for storyboarding, then polished assets later.

Rivas added that users could eventually rely on Gemini Omni, a multimodal large language model that can generate video from images, audio, and text, to turn those prototypes into creative assets. Those capabilities, he said, are still a ways off.

What are the limitations?

Real ones, and Google isn't hiding them. Rivas stressed throughout the briefing that it's still early days for Genie 4 and that the end-to-end product experience remains unclear. The team acknowledged concrete gaps: details and objects can drift the longer a simulation runs, and the researchers haven't yet figured out how to ensure audio stays accurate or in-world text stays legible.

For developers and studios weighing adoption, that means treating Genie 4 as a research preview, not a production tool — at least until the drift and audio problems shrink.

When can you get it?

Genie 4 is available now as a research preview for trusted testers. In the coming weeks, it rolls out to Google AI Ultra subscribers through Project Genie, Google's tool for generating interactive worlds from text or image prompts, and to early access partners of the Gemini API and Google AI Studio. With Runway, AMD, Amazon, and Rivian all pushing into the same territory, the two-minute world is now the benchmark to beat.