Claude Opus 5 · BrowseComp(1,266 题;单代理与多代理按协议区分) 90.8% · 附件证据(ev-op5-card-browsecomp-headline)

Claude Opus 5 的 BrowseComp · 1,266 题;单代理与多代理按协议区分 官方出处。BrowseComp · max effort · 90.8%。见《Claude Opus 5 System Card》System Card PDF pp.152, 162-163, Table 8.1.A and §8.10.2。证据 ev-op5-card-browsecomp-headline。

原文与定位

Claude Opus 5 · BrowseComp(1,266 题;单代理与多代理按协议区分),BrowseComp · max effort配置,Accuracy 90.8%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF pp.152, 162-163, Table 8.1.A and §8.10.2。 查看成绩条件

Table 8.1.A reports BrowseComp at 90.8%. Section 8.10.2 describes web search, web fetch, programmatic tool calling, code execution, context compaction at 200K tokens, and Opus 4.7 grading.
官方材料
Claude Opus 5 System Card · PDF 第 162 页
定位
附件 · System Card PDF pp.152, 162-163, Table 8.1.A and §8.10.2
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 · BrowseComp(1,266 题;单代理与多代理按协议区分),BrowseComp · max effort配置,Accuracy 90.8%(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.152, 162-163, Table 8.1.A and §8.10.2,https://www.anthropic.com/claude-opus-5-system-card#page=162。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-browsecomp-headline/

关联成绩与条件

Claude Opus 5 · BrowseComp 成绩与配置
指标与成绩配置与评测条件证据关系
Accuracy90.8%原文 90.8% · 单位 %
公布 未确认
BrowseComp · max effort
设置
Single-agent BrowseComp with web search, web fetch, programmatic tool calling, and code execution; context compaction triggered at 200K; Opus 4.7 grader; card-wide default max effort.
未知项
未另记
工具
允许工具
推理预算
max
框架
未记录
协议
BrowseComp single-agent with web search, web fetch, programmatic tool calling, and code execution.
条件
90.8%; context compaction triggered at 200K tokens; Opus 4.7 grader; keep separate from the unreleased-effort budget sweep and the 10M fixed-budget effort sweep. Table 8.1.A additionally states default sampling settings, five trials, and evaluation-dependent context windows not exceeding 1M tokens; it does not state this row's per-task token budget.

直接支持该成绩

在排名中查看此成绩