DigestAI news desk

Cut through the AI noise.

Model · multimodal model

Genie 3

Genie 3 is a world model that generates 3D environments with real-time physics from text or image prompts, operating at 720p resolution and 24 frames per second. Recent coverage details its release by Google and the subsequent announcement of Genie 4 by Google DeepMind. The latter introduces photorealistic simulations with native audio, marking an advancement in interactive AI generation.

Genie 3 is a multimodal model from Google, first covered on 27 October 2026. It is available in a consumer app.

Last updated

Quick facts

Who makes Genie 3?
Genie 3 is made by Google. Google news →
When was Genie 3 released?
The first report of Genie 3's release that Digest AI covered was published on 27 October 2026. Google releases Genie 3 world model with 720p, 24fps limits →
What type of model is Genie 3?
Genie 3 is a multimodal model.
Where is Genie 3 available?
Genie 3 is available in a consumer app.

Taken from the coverage, not from our own testing. A question is left out when none of the reports we read gave the answer. All tracked models →

3 stories about Genie 3

  1. Research new

    Google DeepMind releases Genie 4 world model in research preview

    Google DeepMind announced Genie 4, its latest world model for real-time interactive simulations, on October 27, 2026. The model generates photorealistic environments with native audio—including music, ambient sounds,…

    6 sources
    The DecoderTechCrunch AIUnite.AItheverge.com +2 more
  2. Google releases Genie 3 world model with 720p, 24fps limits

    Google upgraded its Genie 3 world model, a 3D environment generator that simulates physics in real time from text or image prompts. The model runs at up to 720p resolution and 24 frames per second, far below gaming…

    1 source
    cryptobriefing.com
  3. Runway says it wants real-time AI video generation as a live stream

    Runway presented research on turning AI video generation into a live, controllable stream. The company says users would describe a scene and see video frames appear instantly, instead of waiting for a completed clip.…

    1 source
    The Decoder
Other models from Google