Claude Fable 5.2 · Humanity's Last Exam(with tools;题集版本未注明) 74.2% · 主文证据(ev-f52-hle-tools)

Claude Fable 5.2 的 Humanity's Last Exam · with tools;题集版本未注明 官方出处。HLE with tools / max · 74.2%。见《Introducing Claude Fable 5.2 and Claude Mythos 5.2》benchmark table > Multidisciplinary reasoning > Humanity's Last Exam (with tools)。证据 ev-f52-hle-tools。

原文与定位

Claude Fable 5.2 · Humanity's Last Exam(with tools;题集版本未注明),HLE with tools / max配置,Reported score 74.2%(具体成绩公布时间未确认)。见《Introducing Claude Fable 5.2 and Claude Mythos 5.2》benchmark table > Multidisciplinary reasoning > Humanity's Last Exam (with tools)。 查看成绩条件

Humanity's Last Exam (with tools) — Fable 5.2: 74.2%; Fable 5.1: 65.6%.
官方材料
Introducing Claude Fable 5.2 and Claude Mythos 5.2 · 定位到章节
定位
主文 · benchmark table > Multidisciplinary reasoning > Humanity's Last Exam (with tools)
材料版本
Official release page dated November 9, 2026; page records an edit on November 11; current HTML/TXT saved as 本地留存副本 and .txt.
版本公布
披露时间
未确认 · 首发/后补未确认
采集时间
历史依据
Same-day CDX HTTP 200 capture 2026-11-09T18:57:13Z, digest WJME3RZDDQECSJSTWPJT463IVXCHMMLP; archived replay was HTTP 200 and is saved as 本地留存副本. Direct card attachment not recovered.
引用文本

Claude Fable 5.2 · Humanity's Last Exam(with tools;题集版本未注明),HLE with tools / max配置,Reported score 74.2%(具体成绩公布时间未确认)。官方出处 Introducing Claude Fable 5.2 and Claude Mythos 5.2(主文,Official release page dated November 9, 2026; page records an edit on November 11; current HTML/TXT saved as 本地留存副本 and .txt.)benchmark table > Multidisciplinary reasoning > Humanity's Last Exam (with tools),https://www.anthropic.com/claude-fable-and-mythos-5-2#evaluations。Benchlog https://benchlog.lekshome.cc/evidence/claude-fable-and-mythos-5-2/ev-f52-hle-tools/

关联成绩与条件

Claude Fable 5.2 · 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
Reported score74.2%原文 74.2% · 单位 %
公布 未确认
HLE with tools / max
设置
Claude Fable 5.2;thinking auto;max effort;web search, web fetch, programmatic tool calling, code execution;980k-token task budget;no context compaction。
未知项
该 effort 点的具体 trial 数未在图注列明。
工具
允许工具
推理预算
max
框架
未记录
协议
Humanity's Last Exam with tools;版本/工具配置未在主文表格列出。
条件
Thinking auto; web search/fetch, programmatic tools and code execution; 980k-token budget; no context compaction; Opus 4.6 grader; max effort.

直接支持该成绩

在排名中查看此成绩