Black Forest Labs Launches FLUX 3, a Joint Image-Video-Audio-Action Model, Plus FLUX-mimic for Robotics
AI video model releases
Sign in to create alerts.
Black Forest Labs Launches FLUX 3, a Joint Image-Video-Audio-Action Model, Plus FLUX-mimic for Robotics
Black Forest Labs released FLUX 3, a multimodal frontier model that jointly trains on images, video, and audio in one architecture, with an extension for action prediction.
FLUX 3 builds on the company's Self-Flow method and ships in four variants: FLUX 3 Video, FLUX 3 Image, FLUX 3 Action, and FLUX 3 Dev.
FLUX 3 Video is still in development but already leads early evaluations against rival frontier video models, standing out on facial expressions, sound-to-event association, and multilingual handling.
Canva, Burda, Magnific (formerly Freepik), Krea, and Picsart are already testing FLUX 3, continuing the FLUX family's presence in tools like Adobe Photoshop and Picsart.
With mimic robotics, Black Forest Labs introduced FLUX-mimic, a video-action model being tested with Audi that can fine-tune to a new manipulation task with as little as 30 minutes of robot data versus 30-plus hours previously.
cho.sh