STEM Professor Bench · 64-problem internal research subset; score averaged over up to nine attempts per problem · 测试定义与成绩

STEM Professor Bench · 64-problem internal research subset; score averaged over up to nine attempts per problem 收录 1 次模型发布的评测披露或成绩。

测试版本与指标

别名 STEM Professor Bench

  • Mean score · % · 越高越好

各系列最新模型 · 公布成绩

常规展示范围。排名按指标与评测协议分组,已记录的工具、预算及框架分别展示。条件未说明的组保留公布值供查证。

Mean score %

工具条件未记录 / max / 框架未记录 · STEM Professor Bench 64-problem subset; max effort; up to nine attempts per problem.

STEM Professor Bench · 64-problem internal research subset; score averaged over up to nine attempts per problem · Mean score · 工具条件未记录 / max / 框架未记录 · STEM Professor Bench 64-problem subset; max effort; up to nine attempts per problem.
模型与配置成绩(%)记录与证据
Claude Fable 5.2Max61.8%

历史记录 · 1 次发布

  • Claude Fable 5.2本次发布有 1 条成绩记录,含 1 条精确值(可含历史记录)