Black Forest Labs Launches FLUX 3, Its First Video Generation Model With Synced Audio
AI video model releases
Black Forest Labs Launches FLUX 3, Its First Video Generation Model With Synced Audio
Black Forest Labs released FLUX 3 in early access, the German AI lab's first model that generates video rather than only still images.
FLUX 3 produces clips up to 20 seconds long with audio generated alongside the video and synced to on-screen action, including dialogue, sound effects, and ambient noise.
The model is multimodal, trained jointly on images, video, and audio within one shared system instead of separate bolted-together tools.
The same backbone powers FLUX-mimic, a robotics model built with mimic robotics, which Audi is already testing on its production line.
Only the open-weight Dev version is planned for later in 2026; the Video and Motion variants stay behind APIs and partner access for now, with an Image version coming in the following weeks.
cho.sh