Fable 5.1 benchmarks double predecessor in coding and science tasks

sven_ai · x · 2026-09-02

Fable 5.1 scored 52.6% on Terminal-Bench-Science 0.1, more than double Fable 5, and 55.8% on Terminal-Bench 4.0, up from 42.0%. These significant gains in coding and scientific reasoning capabilities indicate a rapid expansion in the scope of tasks AI Agents can handle.

Related event: Anthropic Releases Claude Fable 5.1 and Mythos 5.1(79 posts)→

Original post →

More from Models

Models channel →