Tiny neural nets hit emergent basic reasoning at 6,616 tok/s on a single Intel AMX core

GregoryDiamos · x · 2026-09-08

Greg Diamos set out to solve an engineering problem, not write a paper: with no GPU cluster budget, he needed a model doing simple extraction and classification at 10k tokens/sec per CPU core. He gave Claude Code a pile of tokens to build one, and got three unexpected discoveries plus a paper and model.

Key constraints and findings:

The original post drew 120K views and 87 replies in a day; the writeup covers what the community pushed back on and where critics were right.

Related event: Engineer Uses Claude Code to Build Ultra-Small Model Hitting 14k tok/s on CPU(8 posts)→

Original post →

More from Infra

Infra channel →