TiledTensor/TiledCUDA
TiledCUDA is a highly efficient kernel template library designed to elevate CUDA C’s level of abstraction for processing tiles.
C++MIT
Stargazers
- caomwAHU
- CaoWGGnanjing
- constroyShanghai AI Laboratory
- crapromer
- DD-DuDaData Science and Analytic Thrust, Information Hub, HKUST(GZ)
- fly51flyPRIS
- gfvvzShanghai, China
- haozhihanPeking University
- haruhi55Microsoft Research Aisa
- HuyNguyen-hustUniversity of Oregon
- iceplosion
- irasin
- Jason-cs18NYU
- jnulzlGuangZhou China
- jomivaanLisbon
- KAOZUOI
- KipsoraUniversity of Toronto
- KuangjuXUCAS
- learning-chip
- leleucas
- liaoyinanShenzhen, Guangzhou, China
- Light-of-HersPeking University
- luliyucoordinatehangzhou
- lygztqbytedance
- MrGeek-zrhUCAS
- pickteemoBeijing
- qelk123XJTU
- rhmaaa
- smile-luobinHW->MGTV->...
- VAthreePeking University
- YangWang92
- YdrMasterQiYuanLab
- yhwang-hub
- yzh119@flashinfer-ai
- zhangkejiang
- zzzDavid@cornell-zhang