Appgithub-com-thu-ml-sageattention

Download Pinokio

Sign in Create accountJoin

Sign in Create accountJoin

SageAttention

github.com/thu-ml/sageattentionUpdated 9mo ago

[ICLR2025, ICML2025, NeurIPS2025 Spotlight] Quantized Attention achieves speedup of 2-5x compared to FlashAttention, without losing end-to-end metrics across language, image, and video models.

Loading community details…

Loading project owner…

Loading launchers…

Loading check-ins…

Community

Sort byCreate post

Ask or share about SageAttention…

Loading...

Loading references…

Loading project activity…

Pinokio

0

Available launchers88

Ultimate-TTS-StudioFeaturedgithub.com/pinokiofactory/Ultimate-TTS-Studiopip install from git in torch.js62 check-ins · updated 20d ago

DiaFeaturedgithub.com/pinokiofactory/diapip install from git in torch.js0 check-ins · updated 5mo ago

unoFeaturedgithub.com/pinokiofactory/unopip install from git in torch.js0 check-ins · updated 10mo ago

Check-ins (0)

Recent commits

Jintao Zhang9 months agod1a57a5

Update HunyuanVideo example and README

Weiqi-Feng10 months ago5a6f53c

Merge pull request #307 from cooper1637/main

Haoxu Wang10 months agoff71e1e

Restore previous README

whx100310 months ago7608f44

Revert previous reverts as the issue is fixed.

whx100310 months agoc03f15f

Update example inference scripts

jt-zhang10 months ago5061f38

clean revertion

jt-zhang11 months agoc9f4cd7

Revert "Remove torch.compile related tests"

jt-zhang11 months ago06418c2

Revert "Merge pull request #218 from guilhermeleobas/guilhermeleobas/torch-compile"

jt-zhang11 months agoe5bf6ee

Revert "add `mutates_args=("output")`"

jt-zhang11 months ago2fff52e