Nunchux AI Introduces VC-Attention: A Training-Free Low-Bit Attention Kernel That Speeds Up Video Diffusion Transformers

MarkTechPost
Read full post
Nunchux AI launched VC-Attention, a training-free low-bit attention kernel for video Diffusion Transformers that reduces value quantization error and accelerates the softmax stage, cutting generation time significantly.

More in Machine Learning

Microsoft AI CEO criticises Anthropic over model ‘rights’

Covered by 3 sources
Machine Learning4 min read

Noetive launches with $41M to bring self-improving AI to factories and logistics

SiliconANGLE
Machine Learning4 min read

CADDi raises $114M at $1.2B valuation to bring manufacturing AI to North America

SiliconANGLE