Cursor Launches CursorBench 3.2: Fable 5.1 Tops Coding Agent Leaderboard

Cursor has released CursorBench 3.2, a coding agent benchmark based on ambiguous multi-file tasks from real sessions. Fable 5.1 Max leads with 73.4%, while Grok 4.6 follows at 70.8% at roughly a quarter of the price.

2026-09-02 ~ 2026-09-03 · 4 related posts

• Cursor Bench: Grok 4.6 scores 70.8% at a quarter of the leader's price — ChrisGPT · 2026-09-02
• Grok 4.6 and Fable 5.1 lead CursorBench pareto frontier — GavinSBaker · 2026-09-02
• CursorBench 3.2: Fable 5.1 Max Tops, Grok 4.6 Shows Strong Cost Efficiency — mattyp · 2026-09-02

1 near-duplicate retellings: mattyp