Claude Opus 5 · ARC-AGI-3(ARC Prize Foundation semi-private 2026 evaluation set) 30.16% · 附件证据(ev-op5-card-arc-agi-3-2026)

Claude Opus 5 的 ARC-AGI-3 · ARC Prize Foundation semi-private 2026 evaluation set 官方出处。ARC-AGI-3 High effort · 30.16%。见《Claude Opus 5 System Card》System Card PDF pp.186-187, §8.14.2 and Figure 8.14.2.A。证据 ev-op5-card-arc-agi-3-2026。

原文与定位

Claude Opus 5 · ARC-AGI-3(ARC Prize Foundation semi-private 2026 evaluation set),ARC-AGI-3 High effort配置,Relative Human Action Efficiency 30.16%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF pp.186-187, §8.14.2 and Figure 8.14.2.A。 查看成绩条件

The ARC Prize Foundation reports a verified Relative Human Action Efficiency (RHAE) score of 30.16% on the semi-private 2026 evaluation at high effort. Max-effort results were not available at release.
官方材料
Claude Opus 5 System Card · PDF 第 186 页
定位
附件 · System Card PDF pp.186-187, §8.14.2 and Figure 8.14.2.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 · ARC-AGI-3(ARC Prize Foundation semi-private 2026 evaluation set),ARC-AGI-3 High effort配置,Relative Human Action Efficiency 30.16%(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 pp.186-187, §8.14.2 and Figure 8.14.2.A,https://www.anthropic.com/claude-opus-5-system-card#page=186。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-arc-agi-3-2026/

关联成绩与条件

Claude Opus 5 · ARC-AGI-3 成绩与配置
指标与成绩配置与评测条件证据关系
Relative Human Action Efficiency30.16%原文 30.16% · 单位 %
公布 未确认
ARC-AGI-3 High effort
设置
System Card table marks the 30.2% ARC-AGI-3 result as high effort.
未知项
Version and harness details not stated in the table.
工具
工具条件未记录
推理预算
high
框架
未记录
协议
ARC-AGI-3; ARC Prize Foundation semi-private 2026 evaluation; Relative Human Action Efficiency.
条件
High effort; max-effort result was not available at release.

直接支持该成绩

在排名中查看此成绩