Genie 3

Headlines · 3

1. 4d
   Google DeepMind Previews Genie 4 Interactive World Model

   Google DeepMind released Genie 4 as a research preview for trusted testers, with access planned for Google AI Ultra subscribers and early Gemini API and Google AI Studio partners. The world model adds real-time text-driven scene editing, native audio and Gemini-assisted simulation, with scenes reportedly lasting about two minutes, though its realism and two-minute duration are vendor claims and independent testing is limited.

   Google DeepMind Blog · 🔥 30
2. Oct 16
   ByteDance world model demo surfaces, reportedly overseen by Zhang Yiming

   An analysis published by TMTPost describes a Xiaohongshu note in which Tim of the channel 影视飓风 demonstrated ByteDance's latest world model, creating a world and interacting with it — controlling camera movement in old films, stepping through a "door" into imagined scenes, letting robots learn manipulation (including driving) in a virtual world, and using a VR headset for simulated professional training such as medicine. The note named no model, and ByteDance has not officially announced one. Citing unnamed people familiar with the matter, the article says the model is built on ByteDance's video generation model Seedance, generates 20 frames per second on demand with about 50 ms end-to-end latency, renders in the cloud and streams to devices, is planned for Pico headsets, and targets live streaming, interactive short dramas and games. It also says Zhang Yiming personally coordinates model research, cloud infrastructure, Pico hardware and content teams with priority compute, and that world models rank first among ByteDance's four AI priorities for 2026; none of these details has been confirmed by ByteDance.

   钛媒体 · 🔥 0
3. Oct 14
   Reactor adds NVIDIA, Sapphire funding for real-time inference platform

   36Kr reports that startup Reactor is building a real-time inference platform for video and world models, saying developers can use a unified SDK and API to build interactive applications with more than 60 FPS and under 40 milliseconds of latency. Reactor had raised a $59 million seed-plus-Series-A round and recently added follow-on Series A funding from NVIDIA and Sapphire Ventures, bringing its total raised to $74 million. The report also says Reactor worked with an AWS team to run real-time video generation on a single Trainium2 chip, though the performance and deployment claims were not independently measured.

   36氪 人工智能 · 🔥 0

Experience and discussion from the community

Share my Genie 3 experienceAsk about Genie 3