Hacker News new | ask | show | jobs
by nixon_why69 31 days ago
Why not have a bunch of SRAM and various operations like "Q4 matmul" in silicon? Model weights and even architectures could still evolve on a platform like that.
2 comments

Doesnt "a bunch of SRAM" top out at maybe a few gigs per chip (with zero area used for logic)? You'd need an order of magnitude more to fit even a fairly weak general purpose LLM model.
I belive that is what NPUs are.

The issue is the very huge amount of DRAM and high bandwidth these model require.