MiniMaxAnalysisAI ModelsSeptember 15, 2026

Nunchux AI's VC-Attention speeds up video diffusion transformers

VC-Attention is a training-free low-bit attention kernel for video Diffusion Transformers that targets value quantization error and a slow softmax stage. MiniMax highlighted it as one of three paths to faster video attention.

How this story unfolded

1 day · 2 reports · 2 community posts · from Sep 17

  1. Sep 17
  2. Sep 18

More stories today

Open the live feed