B200 Attention Kernel from Scratch to Near-SOTA in 60 Diagrams | Hacker News

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

		B200 Attention Kernel from Scratch to Near-SOTA in 60 Diagrams (iaroslavelistratov.github.io)
		23 points by magoghm 1 day ago \| hide \| past \| favorite \| 3 comments

		help

xiphias2 1 day ago | [–]

Pretty cool, it's missing how these concepts map to ThunderKittens code (my favorite CUDA toolkit)

peter_d_sherman 20 hours ago | | [–]

One of the best AI GPU Attention Kernel how-to-implement articles I've read in a long time!

Tremendous work, tremendous effort was placed into writing this article -- and it shows!

Well done!

erichocean 1 day ago | [–]

Really interesting project, appreciated.

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