Claude Opus 5 · Global MMLU(42 languages including English · overall average) 92.5% · 附件证据(ev-op5-card-gmmlu)

Claude Opus 5 的 Global MMLU · 42 languages including English · overall average 官方出处。GMMLU · max, single trial · 92.5%。见《Claude Opus 5 System Card》System Card PDF p.190, Figure 8.16.1.A。证据 ev-op5-card-gmmlu。

原文与定位

Claude Opus 5 · Global MMLU(42 languages including English · overall average),GMMLU · max, single trial配置,Average accuracy 92.5%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.190, Figure 8.16.1.A。 查看成绩条件

Visual transcription (Claude Opus 5 chart label): 92.5% average accuracy. Caption: max effort; single trial; no tools or customized system prompt.
官方材料
Claude Opus 5 System Card · PDF 第 190 页
定位
附件 · System Card PDF p.190, Figure 8.16.1.A
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 · Global MMLU(42 languages including English · overall average),GMMLU · max, single trial配置,Average accuracy 92.5%(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.190, Figure 8.16.1.A,https://www.anthropic.com/claude-opus-5-system-card#page=190。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-gmmlu/

关联成绩与条件

Claude Opus 5 · Global MMLU 成绩与配置
指标与成绩配置与评测条件证据关系
Average accuracy92.5%原文 92.5% · 单位 %
公布 未确认
GMMLU · max, single trial
设置
Max effort; no tools or customized system prompt; single trial; 42-language overall average.
未知项
未另记
工具
禁止工具
推理预算
max
框架
未记录
协议
GMMLU 42-language overall average accuracy.
条件
Max effort; no tools or customized system prompt; single trial.

直接支持该成绩

在排名中查看此成绩