TellWorldAI World Archive

Home›News›Release

ReleaseGenerative AI for everyone

Google DeepMind releases Genie 4 in Project Genie and the Gemini API

On 18 November 2026 Google DeepMind announced Genie 4 in Project Genie, then on 19 November made it available to developers in the Gemini API and Google AI Studio. Genie 3 also got longer sessions the same day.

A translucent glass cube holds a miniature landscape of hills, river and clouds that spills beyond its edges.

Background

Genie is Google DeepMind's line of world models, which generate interactive virtual environments from prompts. The 18 November 2026 announcement came together with an update to Genie 3.[1]

What happened

On 18 November 2026 Google DeepMind announced Genie 4 and rolled it out to eligible Google AI Ultra subscribers in Project Genie.[1] The same day, Google extended Genie 3 sessions to up to 3 minutes per generation.[1]

On 19 November 2026 Genie 4 became available to developers in the Gemini API and Google AI Studio.[2]

Key details

Google said the API version runs through the Interactions API and starts from a text prompt, a reference image or both.[2] Developers can navigate generated worlds with real-time keyboard controls, and they can change weather, lighting, objects or events with live text prompts.[2] Google also said Gemini reasoning keeps the world state consistent throughout a session.[2]

Impact

The release gave the same model two audiences: subscribers using Project Genie, and developers building through the Gemini API.[1][2]

Terms used in this entry

An API is the interface through which a developer sends requests to a model and receives its output, so the same model can be used inside other software. Google AI Studio is the web tool where developers can try Gemini models before building with them. A text prompt is a written description of what the model should produce, and a reference image gives the model a visual starting point to build from.

Developers and subscribers reach the model through different routes. Project Genie is an experience for people to explore generated worlds directly, while the Gemini API is for software built by developers. A developer who uses the API decides how the output is shown, stored or combined with other features, so the same model can serve very different products.

Sources

  1. 1
  2. 2

Cite this page: Google DeepMind releases Genie 4 in Project Genie and the Gemini API. TellWorld, https://www.tellworld.net/news/genie-4-2026. Updated 2026-11-20

Related news

View all