VC-Attention

Mentions · 7d0
Mentions · 30d0
All mentions2
First seenSep 13, 2026
Last mentionedSep 17, 2026

Timeline

2 articles · newest first

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

Nunchux AI introduces VC-Attention, a training-free low-bit attention kernel that speeds up video diffusion transformers up to 3.58x.

MarkTechPostSep 17, 00:45 UTC · Sep 17, 2026AI research6

VC-Attention: Value Smoothing and Softmax Casting for Low-bit Attention

VC-Attention is a training-free low-bit attention method for diffusion transformers, achieving 1.46-1.59x kernel speedups on datacenter GPUs with higher fidelity.

Hugging Face daily papersSep 13, 20:00 UTC · Sep 13, 2026AI research3 · 1 read

Appears with

LongCat-Videoai model · 2
MiniMax-H3ai model · 2
HunyuanVideo-1.5ai model · 2
Wan2.2ai model · 1
Wan2.2-T2V-A14Bai model · 1
NVIDIAvendor · 1
FlashAttention-4product · 1
V-Smoothproduct · 1
SageAttention3product · 1
Nunchux AIvendor · 1
SageAttention2product · 1
ExpCast-FP8product · 1

Entities are extracted by the model from each article. First seen Sep 13, 2026, last mentioned Sep 17, 2026. Watching keeps it in this browser only (no account); the watchlist page and dashboard alerts use it.