Home Model series Claude Fable Claude Fable 5.1 Claude Fable 5.1 · Terminal-Bench-Science · 0.1 · Scores and configurations Claude Fable 5.1 · Terminal-Bench-Science · 0.1 · Scores and configurations Claude Fable 5.1 record for Terminal-Bench-Science · 0.1. 1 score record for this release, including 1 exact value (may include historical records). Includes 2 source locations.
Scores and evaluation conditions Claude Fable 5.1 · Terminal-Bench-Science scores and configurations Metric and score Configuration and evaluation conditions Sources Score 52.6% Reported 52.6% · Unit % Published Unconfirmed
Adaptive Max Settings Claude Fable 5.1;CursorBench 主文引述明确为 max effort;后续 System Card 对 OSWorld/BenchCAD/OfficeQA 对照结果亦标 max effort。 Unknowns 不同 benchmark 的具体执行细节分别见来源章节;部分测试的采样参数或运行数未单独公布。 Tools Tool access not recorded Reasoning budget maximum Framework Claude Code --bare mode Protocol Terminal-Bench-Science 0.1; 70 tasks; Claude Code bare mode; maximum thinking effort; 10 trials per task (700 trials). Conditions Fable 5.1 System Card §8.7 (PDF p.171) confirms Claude Code --bare mode, maximum thinking effort, and 10 trials per task (700 trials). View this score in rankings
Benchmark versions and metrics Aliases Terminal-Bench-Science 0.1
Score · % · Higher is better