# @myhandle — 2026-08-19

♥1 ↻0 · https://x.com/myhandle/status/2090138038630486192

@davidmanheim @pangram the bitter lesson says that Mantic will train a better forecaster if they trained an even bigger model than gpt-oss-120B on even more events rather that hard-coding heuristics. it doesn't say that Grok 4.5 is a better forecaster because it has 10x the parameters (and it isn't)

tags: author:myhandle, kind:tweet, on:grok-4-5, year:2026
cited on: _dossiers/grok-4-5.md, grok-4-5