Black Forest Labs launches FLUX 3 multimodal foundation model

Black Forest Labs released FLUX 3, a unified multimodal model jointly trained on images, video and audio under a Self-Flow architecture. It generates video with native audio up to 20 seconds, supports image synthesis and editing, and extends to action prediction for physical AI applications. Early access for video is available, with open-weight FLUX 3 Dev planned later and early robotics deployment already underway.