Home Model series Claude Fable Claude Fable · ProgramBench · 166 个保留的 golden tasks;架构与时限按协议区分 · Release history Claude Fable · ProgramBench · 166 个保留的 golden tasks;架构与时限按协议区分 · Release history View scores, configurations, and disclosures for ProgramBench · 166 个保留的 golden tasks;架构与时限按协议区分 across 2 Claude Fable releases, ordered by release date.
Differences are calculated only for unique, active, exact values between direct predecessors when the unit, metric, and explicitly recorded configuration and conditions match.
2026-09-15 · 1 score record for this release, including 1 exact value (may include historical records)
Claude Fable 5.1 · ProgramBench scores and configurations Metric and score Configuration and evaluation conditions Sources Hidden test pass rate 87.6% Reported 87.6% · 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 max Framework mini-SWE-agent harness Protocol ProgramBench; 166 golden tasks after excluding 34 tasks with flaky reference binaries; same mini-swe-agent harness without its six-hour time limit. Conditions Single-agent result; multi-agent curves are nonstandard harness experiments and are not ranked. View this score in rankings
本地留存副本, PDF p.176 Attachment · Claude Fable 5.1 and Claude Mythos 5.1 System Card 2026-11-09 · 2 score records for this release, including 2 exact values (may include historical records)
Records for the direct predecessor do not provide matching conditions for a valid difference.
Claude Fable 5.2 · ProgramBench scores and configurations Metric and score Configuration and evaluation conditions Sources Hidden test pass rate 85.6% Reported 85.6% · Unit % Published Unconfirmed
ProgramBench without remaining-context display Settings Claude Fable 5.2;single-agent mini-swe-agent harness;1M-token context;remaining context tokens are not shown;max effort。 Unknowns none recorded Tools Tool access not recorded Reasoning budget max Framework single-agent mini-SWE-agent harness Protocol ProgramBench 166-task golden subset; single-agent mini-swe-agent harness; no internet/decompilation; hidden-test pass rate. Conditions 1M-token context with no remaining-context indicator; max effort. Hidden test pass rate 91.6% Reported 91.6% · Unit % Published Unconfirmed
ProgramBench with remaining-context display Settings Claude Fable 5.2;single-agent mini-swe-agent harness;1M-token context;remaining context tokens are shown after each step;max effort。 Unknowns none recorded Tools Tool access not recorded Reasoning budget max Framework single-agent mini-SWE-agent harness Protocol ProgramBench 166-task golden subset; single-agent mini-swe-agent harness; no internet/decompilation; hidden-test pass rate. Conditions 1M-token context with remaining-context tokens shown after each step; max effort.
PDF pp.215-216, §8.10.1, ProgramBench Attachment · Claude Fable 5.2 and Claude Mythos 5.2 System Card