FLUX 3 Action builds on the same image, video, and audio pretraining as FLUX 3, but uses a smaller architecture. More efficient representations learned through our research on Self-Flow made this smaller size possible. In midtraining, we trained the model to predict actions and future frames together.
FLUX 3 Action uses a smaller architecture to predict actions and frames
AISummary
Black Forest Labs says FLUX 3 Action builds on the same image, video, and audio pretraining as FLUX 3 but uses a smaller architecture. The company attributes this smaller size to more efficient representations learned through its Self-Flow research. During midtraining, the model was trained to predict actions and future frames together.
Post on XView on X
Source: Black Forest Labs · x.comPublished · added here