KV Sharing, MHC, and Compressed Attention

(magazine.sebastianraschka.com)

24 points | by gmays 5 hours ago ago

2 comments

  • nibab 2 hours ago ago

    cool stuff. my comp sci major feels almost completely redundant in this new vibecoding era and i feel like the only way to stay relevant as a programmer is to learn all these compute primitives and become an LLM systems guy.

  • redwood 22 minutes ago ago

    Has anyone seen a similar deep dive but that looks a little bit more closely at the infrastructure building blocks that power each of the components. I mean something a bit more physically grounded like how much compute would go to each portion to serve a Frontier Model?