Claude Opus 5 · International Mathematical Olympiad(2026 contest; six proof problems) 42 points · 附件证据(ev-op5-card-imo-2026)

Claude Opus 5 的 International Mathematical Olympiad · 2026 contest; six proof problems 官方出处。IMO 2026 · proof evaluation · 42 points。见《Claude Opus 5 System Card》System Card PDF p.157, §8.6 IMO 2026。证据 ev-op5-card-imo-2026。

原文与定位

Claude Opus 5 · International Mathematical Olympiad(2026 contest; six proof problems),IMO 2026 · proof evaluation配置,Score 42 points(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.157, §8.6 IMO 2026。 查看成绩条件

All 24 independently generated solutions were judged correct; the final score is 42/42, corresponding to gold-medal performance. Human experts graded one pre-specified solution per problem at 7/7.
官方材料
Claude Opus 5 System Card · PDF 第 157 页
定位
附件 · System Card PDF p.157, §8.6 IMO 2026
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 · International Mathematical Olympiad(2026 contest; six proof problems),IMO 2026 · proof evaluation配置,Score 42 points(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.157, §8.6 IMO 2026,https://www.anthropic.com/claude-opus-5-system-card#page=157。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-imo-2026/

关联成绩与条件

Claude Opus 5 · International Mathematical Olympiad 成绩与配置
指标与成绩配置与评测条件证据关系
Score42 points原文 42/42 · 单位 points
公布 未确认
IMO 2026 · proof evaluation
设置
Adaptive thinking at max effort; no tools or agent harness; 256,000-token output limit; four independent solutions per problem. If the output limit was exhausted, attempts were resampled at lower thinking effort; one solution required this.
未知项
The final 42/42 score aggregates solutions with this stated resampling exception; it is not a uniform single-effort API run.
工具
禁止工具
推理预算
输出上限 256000 tokens
框架
未记录
协议
IMO 2026 six proof problems; three-model judge panel; one pre-specified solution per problem also human graded.
条件
Four independent solutions per problem; 24/24 panel judgments correct; human experts graded the first solution packet at 7/7 per problem. Output-limit exhaustion caused one lower-effort resample.

直接支持该成绩

在排名中查看此成绩