@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)
cited on: grok-4-5
Reproduced against link rot, credited and linked to its original. Yours and you’d rather it weren’t here? Open an issue.