ReHyAt: Recurrent Hybrid Attention for Video Diffusion Transformers

Open paper
Benchmark Model Rank Results
text-to-video-generation-on-vbenchWan2.1 1.3B + ReHyAt (15×Tc=3, To=1)–VBench Total Score: 83.79
text-to-video-generation-on-vbench-2-0ReHyAt (15×Tc=5)–VBench-2.0 Total Score: 56.3

Rank counts only results with a code link.