Flux Action predicts chunks of 9 to 21 future video frames and action states conditioned on prior visual context and text instructions

Activity

Activity (0)
Loading comments...
Opening sign-in…