Claude Sonnet 5.5 · Global MMLU(42 languages including English · overall average) 92.1% · 附件证据(ev55-gmmlu)

Claude Sonnet 5.5 的 Global MMLU · 42 languages including English · overall average 官方出处。Adaptive thinking, max effort · 92.1%。见《Claude Sonnet 5.5 System Card》§8.16.1 Figure 8.16.1.A, PDF p.140。证据 ev55-gmmlu。

原文与定位

Claude Sonnet 5.5 · Global MMLU(42 languages including English · overall average),Adaptive thinking, max effort配置,Average accuracy 92.1%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.16.1 Figure 8.16.1.A, PDF p.140。 查看成绩条件

8.16.1 GMMLU results On GMMLU, Claude Sonnet 5.5 achieved an average accuracy of 92.1% across 42 languages, behind Claude Opus 5.5 at 94.3% and Claude Fable 5.1 at 94.0%, and ahead of Claude Sonnet 5 at 89.2%.
官方材料
Claude Sonnet 5.5 System Card · PDF 第 140 页
定位
附件 · §8.16.1 Figure 8.16.1.A, PDF p.140
材料版本
PDF cover dated 2026-09-28; 148 pages; collected 2026-11-10
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 Sonnet 5.5 · Global MMLU(42 languages including English · overall average),Adaptive thinking, max effort配置,Average accuracy 92.1%(具体成绩公布时间未确认)。官方出处 Claude Sonnet 5.5 System Card(附件,PDF cover dated 2026-09-28; 148 pages; collected 2026-11-10)§8.16.1 Figure 8.16.1.A, PDF p.140,https://www.anthropic.com/claude-sonnet-5-5-system-card#page=140。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-5-5/ev55-gmmlu/

关联成绩与条件

Claude Sonnet 5.5 · Global MMLU 成绩与配置
指标与成绩配置与评测条件证据关系
Average accuracy92.1%原文 92.1% · 单位 %
公布 未确认
Adaptive thinking, max effort
设置
Adaptive thinking enabled; max effort.
未知项
Sampling settings may vary by benchmark; see each result.conditions.
工具
禁止工具
推理预算
max
框架
未记录
协议
GMMLU average accuracy across 42 languages; one trial.
条件
Adaptive thinking and max effort; no tools or customized system prompt.

直接支持该成绩

在排名中查看此成绩