Modular beat Nvidia's cuBLAS kernels on B200s in 170 LOC | Hacker News

Hacker Newsnew | past | comments | ask | show | jobs | submit

		Modular beat Nvidia's cuBLAS kernels on B200s in 170 LOC (twitter.com/aliestaha)
		3 points by pbd 4 months ago \| hide \| past \| favorite \| 1 comment

pbd 4 months ago [–]

kernel optimisation is paying $$$

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact