Fable 5.1 Tops Bug Hunt Bench, Outperforming GPT-5.6

PawelHuryn · x · 2026-09-02

PawelHuryn released results for the Bug Hunt benchmark, which involves 2 real repos with 105 planted bugs. Fable 5.1 (max) ranked first with 43 bugs fixed, beating GPT-5.6 Sol (max) at 42 and Grok 4.6 at 27. Fable 5.1 is also 2.24x faster than GPT-5.6 Sol and 26% cheaper than the previous Fable 5 model.

Related event: Anthropic's Fable 5.1 Tops Bug Hunt Bench, Beating GPT-5.6(4 posts)→

Original post →

More from coding & agent

coding & agent channel →