Home Model series Claude Fable Claude Fable 5.1 Claude Fable 5.1 · OSWorld · 2.1, 82-task offline no-internet subset, partial score · Scores and configurations Claude Fable 5.1 · OSWorld · 2.1, 82-task offline no-internet subset, partial score · Scores and configurations Claude Fable 5.1 record for OSWorld · 2.1, 82-task offline no-internet subset, partial score. 2 score records for this release, including 2 exact values (may include historical records). Includes 1 source location.
Scores and evaluation conditions Claude Fable 5.1 · OSWorld scores and configurations Metric and score Configuration and evaluation conditions Sources Partial score 83.5% Reported 83.5% partial · 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 max Framework not recorded Protocol OSWorld 2.1 official 82-task offline set;partial score;Pass@1,五次尝试平均。 Conditions Same no-internet task subset and scoring rules as the Fable 5.2 result. View this score in rankings Strict pass rate 47.6% Reported 47.6% strict pass · 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 max Framework not recorded Protocol OSWorld 2.1 official 82-task offline set;strict pass rate;Pass@1,五次尝试平均。 Conditions Same no-internet task subset and scoring rules as the Fable 5.2 result. View this score in rankings
Benchmark versions and metrics Aliases OSWorld-2.1 offline subset
Partial score · % · Higher is better Strict pass rate · % · Higher is better