Remix.run Logo
B200 Attention Kernel from Scratch to Near-SOTA in 60 Diagrams(iaroslavelistratov.github.io)
24 points by magoghm 2 days ago | 3 comments
xiphias2 2 days ago | parent | next [-]

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

peter_d_sherman a day ago | parent | prev | next [-]

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 2 days ago | parent | prev [-]

Really interesting project, appreciated.