Digest · past week

1. Black Forest Labs releases FLUX 3: multimodal model for images, video, audio, and robot action prediction
   Black Forest Labs (BFL) introduces FLUX 3, a multimodal foundation model trained jointly on images, video, audio, and robot action prediction. The model leverages Self-Flow architecture to align modalities, and early tests show strong preference over competitors like Luma Ray 3.2 and Runway Gen-4.5 in text-to-video generation.