Claude Opus 5 · BioMysteryBench(官方发布表列 Hard 与 human-solved 结果;细分版本未注明) · 附件证据(ev-op5-card-biomystery)

Claude Opus 5 的 BioMysteryBench · 官方发布表列 Hard 与 human-solved 结果;细分版本未注明 官方出处。见《Claude Opus 5 System Card》System Card PDF p.192, §8.17.1 BioMysteryBench。证据 ev-op5-card-biomystery。

原文与定位

Claude Opus 5 · BioMysteryBench(官方发布表列 Hard 与 human-solved 结果;细分版本未注明),BioMysteryBench · max配置,Human Difficult subset score 49.4%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.192, §8.17.1 BioMysteryBench。 查看成绩条件

Claude Opus 5 · BioMysteryBench(官方发布表列 Hard 与 human-solved 结果;细分版本未注明),BioMysteryBench · max配置,Human Solvable subset score 90.1%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.192, §8.17.1 BioMysteryBench。 查看成绩条件

Claude Opus 5 scored 90.1% on the Human Solvable subset and 49.4% on the Human Difficult subset. Three Human Solvable and six Human Difficult problems were removed after external feedback.
官方材料
Claude Opus 5 System Card · PDF 第 192 页
定位
附件 · System Card PDF p.192, §8.17.1 BioMysteryBench
材料版本
Current saved PDF/TXT: 本地留存副本 and current-opus-5-system-card.txt; cover dated 2026-07-24; p.2 changelog records an update on 2026-08-19; later revision date, if any, unknown.
版本公布
未确认
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
暂无可靠历史版本依据
引用文本

Claude Opus 5 · BioMysteryBench(官方发布表列 Hard 与 human-solved 结果;细分版本未注明),BioMysteryBench · max配置,Human Difficult subset score 49.4%(具体成绩公布时间未确认);Claude Opus 5 · BioMysteryBench(官方发布表列 Hard 与 human-solved 结果;细分版本未注明),BioMysteryBench · max配置,Human Solvable subset score 90.1%(具体成绩公布时间未确认)。官方出处 Claude Opus 5 System Card(附件,Current saved PDF/TXT: 本地留存副本 and current-opus-5-system-card.txt; cover dated 2026-07-24; p.2 changelog records an update on 2026-08-19; later revision date, if any, unknown.)System Card PDF p.192, §8.17.1 BioMysteryBench,https://www.anthropic.com/claude-opus-5-system-card#page=192。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-biomystery/

关联成绩与条件

Claude Opus 5 · BioMysteryBench 成绩与配置
指标与成绩配置与评测条件证据关系
Human Difficult subset score49.4%原文 49.4% (Human Difficult) · 单位 %
公布 未确认
BioMysteryBench · max
设置
Card-wide default max effort; bash, file editor, and package managers; Human Solvable and Human Difficult subsets; multiple tasks are nonpublic/internal evaluations.
未知项
未另记
工具
允许工具
推理预算
max
框架
未记录
协议
BioMysteryBench Human Difficult subset.
条件
Card-wide default max effort; bash, file editor and package managers; six problems removed after external feedback.

直接支持该成绩

在排名中查看此成绩
Human Solvable subset score90.1%原文 90.1% (Human Solvable) · 单位 %
公布 未确认
BioMysteryBench · max
设置
Card-wide default max effort; bash, file editor, and package managers; Human Solvable and Human Difficult subsets; multiple tasks are nonpublic/internal evaluations.
未知项
未另记
工具
允许工具
推理预算
max
框架
未记录
协议
BioMysteryBench Human Solvable subset.
条件
Card-wide default max effort; bash, file editor and package managers; three problems removed after external feedback.

直接支持该成绩

在排名中查看此成绩