Claude Opus 4.5 · ARC-AGI-1(ARC Prize Foundation private dataset) 80% · 附件证据(ev-op45-card-arc1-private)

Claude Opus 4.5 的 ARC-AGI-1 · ARC Prize Foundation private dataset 官方出处。ARC Prize Foundation private validation, 64K thinking · 80%。见《Claude Opus 4.5 System Card》PDF p.28, §2.10, ARC-AGI results paragraph。证据 ev-op45-card-arc1-private。

原文与定位

Claude Opus 4.5 · ARC-AGI-1(ARC Prize Foundation private dataset),ARC Prize Foundation private validation, 64K thinking配置,Accuracy 80%(具体成绩公布时间未确认)。见《Claude Opus 4.5 System Card》PDF p.28, §2.10, ARC-AGI results paragraph。 查看成绩条件

The ARC Prize Foundation reports that Claude Opus 4.5 achieved 80.0% on ARC-AGI-1 with 64k thinking tokens on their private dataset.
官方材料
Claude Opus 4.5 System Card · PDF 第 28 页
定位
附件 · PDF p.28, §2.10, ARC-AGI results paragraph
材料版本
Current saved PDF/TXT: 本地留存副本 and .txt; cover November 2025, change log 2025-11-24; current binary is not established as launch-day bytes; fully reviewed 2026-11-16.
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 4.5 · ARC-AGI-1(ARC Prize Foundation private dataset),ARC Prize Foundation private validation, 64K thinking配置,Accuracy 80%(具体成绩公布时间未确认)。官方出处 Claude Opus 4.5 System Card(附件,Current saved PDF/TXT: 本地留存副本 and .txt; cover November 2025, change log 2025-11-24; current binary is not established as launch-day bytes; fully reviewed 2026-11-16.)PDF p.28, §2.10, ARC-AGI results paragraph,https://www.anthropic.com/claude-opus-4-5-system-card#page=28。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-4-5/ev-op45-card-arc1-private/

关联成绩与条件

Claude Opus 4.5 · ARC-AGI-1 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy80%原文 80.0% · 单位 %
公布 未确认
ARC Prize Foundation private validation, 64K thinking
设置
64K thinking tokens; ARC Prize Foundation private validation dataset.
未知项
Run count and evaluation harness details are not stated in the card.
工具
工具条件未记录
推理预算
思考预算 64K
框架
未记录
协议
ARC-AGI-1 private validation dataset; ARC Prize Foundation reported result.
条件
64K thinking tokens; the system card says the model was trained on the public ARC-AGI-1 dataset.

直接支持该成绩

在排名中查看此成绩