Bug Hunt Bench: Fable 5.2 Tops Ranking, Fixing 58/105 Real Bugs

Pawel Huryn's Bug Hunt Bench tested frontier coding models on 105 real bugs from two repos: Fable 5.2 (max) topped the leaderboard by fixing 58 of them, while DeepSeek V4.1 Flash dominated on fixes-per-dollar efficiency.

2026-11-10 ~ 2026-11-10 · 3 related posts

• Bug Hunt Bench: Fable 5.2 fixes 58/105 real bugs, beating Sonnet 5.5 at half the cost and turns — PawelHuryn · 2026-11-10
• Bug Hunt Bench full leaderboard: DeepSeek V4.1 Flash fixes 42.7 bugs for under $2 — PawelHuryn · 2026-11-10
• Bug Hunt stats: Fable 5.2 found just 1 never-before-fixed bug across 300+ runs — PawelHuryn · 2026-11-10