Mathematician: latest models Sol 5.6 and Fable 5.1 can't solve any of my real problems

burny_tech · x · 2026-09-07

Pushing back on the "math is dead" takes, a researcher notes how many ordinary math problems AI still fails. Testing Sol 5.6 and Fable 5.1 on problems he has genuinely been thinking about, neither model solved a single one—a counterpoint to headlines of models acing benchmark math.

Original post →

More from Models

Models channel →