Black Forest Labs releases FLUX 3, extended to video and robot control

Black Forest Labs released FLUX 3, a multimodal model trained on images, video, and audio, and mimic built FLUX-mimic on its video backbone to control robots.

In a soft-body kitting task, mimic reports a 95% success rate without single-task fine-tuning, compared with 55% for an adapted π0.5 model.

FLUX 3 Video is in early access, with action prediction offered to selected partners and an open-weight backbone planned.