Claude Opus 4.5 · Humanity's Last Exam(2,500-question benchmark) · 附件证据(ev-op45-card-hle)

Claude Opus 4.5 的 Humanity's Last Exam · 2,500-question benchmark 官方出处。见《Claude Opus 4.5 System Card》PDF pp.31-32, §2.16 and Figure 2.16.A。证据 ev-op45-card-hle。

原文与定位

Claude Opus 4.5 · Humanity's Last Exam(2,500-question benchmark),HLE reasoning-only, no tools配置,Accuracy 30.8%(具体成绩公布时间未确认)。见《Claude Opus 4.5 System Card》PDF pp.31-32, §2.16 and Figure 2.16.A。 查看成绩条件

Claude Opus 4.5 · Humanity's Last Exam(2,500-question benchmark),HLE tools-only, no reasoning配置,Accuracy 43.2%(具体成绩公布时间未确认)。见《Claude Opus 4.5 System Card》PDF pp.31-32, §2.16 and Figure 2.16.A。 查看成绩条件

Claude Opus 4.5: 30.8% without search; 43.2% with search. The two configurations are reasoning-only without tools, and tools-only with web search, web fetch and code execution but no reasoning.
官方材料
Claude Opus 4.5 System Card · PDF 第 32 页
定位
附件 · PDF pp.31-32, §2.16 and Figure 2.16.A
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 · Humanity's Last Exam(2,500-question benchmark),HLE reasoning-only, no tools配置,Accuracy 30.8%(具体成绩公布时间未确认);Claude Opus 4.5 · Humanity's Last Exam(2,500-question benchmark),HLE tools-only, no reasoning配置,Accuracy 43.2%(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 pp.31-32, §2.16 and Figure 2.16.A,https://www.anthropic.com/claude-opus-4-5-system-card#page=32。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-4-5/ev-op45-card-hle/

关联成绩与条件

Claude Opus 4.5 · Humanity's Last Exam 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy30.8%原文 30.8% · 单位 %
公布 未确认
HLE reasoning-only, no tools
设置
Reasoning-only evaluation without tools; 2,500-question HLE benchmark.
未知项
Reasoning budget, trial count and other sampling settings are not stated.
工具
禁止工具
推理预算
未记录
框架
未记录
协议
Humanity's Last Exam, 2,500 questions; reasoning-only, no tools; Claude Sonnet 4.5 grader.
条件
Reasoning budget and trial count not stated.

直接支持该成绩

在排名中查看此成绩
Accuracy43.2%原文 43.2% · 单位 %
公布 未确认
HLE tools-only, no reasoning
设置
Web search, web fetch and code execution; no reasoning.
未知项
Trial count and detailed tool budgets are not stated.
工具
允许工具
推理预算
未记录
框架
未记录
协议
Humanity's Last Exam, 2,500 questions; web search, web fetch and code execution; no reasoning; Claude Sonnet 4.5 grader.
条件
Tool budgets and trial count not stated. The search-enabled arm used transcript checks and manual review to flag likely online answer retrieval.

直接支持该成绩

在排名中查看此成绩