Claude Fable 5.1 Scores Leak: 77.9% on OSWorld, Beating GPT-5.6 Sol Across Benchmarks

Scobleizer · x · 2026-09-02

Leaked benchmark numbers show Claude Fable 5.1 scoring 52.6% on agentic scientific research, 55.8% on Terminal-Bench, 77.9% on OSWorld, and 73.4% on CursorBench — beating both Fable 5 and GPT-5.6 Sol across multiple benchmarks.

Related event: Claude Fable 5.1 Benchmarks Leaked, Crushing Rivals(2 posts)→

Original post →

More from Models

Models channel →