option
Home
Flash News
Content
NicholasGonzález
NicholasGonzález
July 24, 2026

German AI startup Black Forest Labs released Flux3, a multimodal foundation model with Self-Flow architecture integrating image, video, audio, and motion codecs. It supports native audio-video sync up to 20 seconds. In tests, Flux3 outperformed Luma Ray3.2 (93% win rate) and Runway Gen-4.5 (77%). A robotics action model Flux-mimic is being tested at an Audi factory. Flux3Video is live; Flux3Image and open-source Flux3Dev coming soon.

German AI startup Black Forest Labs released Flux3, a multimodal foundation model with Self-Flow architecture integrating image, video, audio, and motion codecs. It supports native audio-video sync up to 20 seconds. In tests, Flux3 outperformed Luma Ray3.2 (93% win rate) and Runway Gen-4.5 (77%). A robotics action model Flux-mimic is being tested at an Audi factory. Flux3Video is live; Flux3Image and open-source Flux3Dev coming soon.
Comments (0)
0/300
OR