Claude Opus 4.7 · OSWorld-Verified(Verified benchmark; evaluation protocol noted per source) 77.9% · 附件证据(ev-op47-card-osworld-chart)

Claude Opus 4.7 的 OSWorld-Verified · Verified benchmark; evaluation protocol noted per source 官方出处。发布表报告值 · 77.9%。见《Claude Opus 4.7 System Card》Figure 8.9.4.A, PDF p.208。证据 ev-op47-card-osworld-chart。

原文与定位

Claude Opus 4.7 · OSWorld-Verified(Verified benchmark; evaluation protocol noted per source),发布表报告值配置,First-attempt success rate 77.9%(具体成绩公布时间未确认)。见《Claude Opus 4.7 System Card》Figure 8.9.4.A, PDF p.208。 查看成绩条件

The Opus 4.7 bar in OSWorld-Verified across Claude models is explicitly labeled 77.9%.
官方材料
Claude Opus 4.7 System Card · PDF 第 208 页
定位
附件 · Figure 8.9.4.A, PDF p.208
材料版本
Current saved PDF/TXT: 本地留存副本 and current-opus-4-7-system-card.txt; 232 PDF pages; metadata created/modified 2026-04-16 22:11:48 China Standard Time. Full current-card review completed 2026-11-16; this file is not established as launch-day bytes.
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 4.7 · OSWorld-Verified(Verified benchmark; evaluation protocol noted per source),发布表报告值配置,First-attempt success rate 77.9%(具体成绩公布时间未确认)。官方出处 Claude Opus 4.7 System Card(附件,Current saved PDF/TXT: 本地留存副本 and current-opus-4-7-system-card.txt; 232 PDF pages; metadata created/modified 2026-04-16 22:11:48 China Standard Time. Full current-card review completed 2026-11-16; this file is not established as launch-day bytes.)Figure 8.9.4.A, PDF p.208,https://www.anthropic.com/claude-opus-4-7-system-card#page=208。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-4-7/ev-op47-card-osworld-chart/

关联成绩与条件

Claude Opus 4.7 · OSWorld-Verified 成绩与配置
指标与成绩配置与评测条件证据关系
First-attempt success rate78%原文 78.0% · 单位 %
公布 未确认

已更正 · 82.3%。官方后续脚注说明评测方式有变并更新旧值;同文图表另列 82.8%,故保留冲突。

结果冲突待核对,精确排名不采用。

发布表报告值
设置
System Card §8.1 standard configuration: adaptive thinking at max effort; default temperature/top_p; averaged over 5 trials. Context sizes are evaluation-dependent and do not exceed 1M tokens; retain all section-specific exceptions on their results.
未知项
Per-benchmark tools, context/token cap, resolution, trial count, harness and other exceptions are recorded on each result when stated.
工具
工具条件未记录
推理预算
max
框架
未记录
协议
OSWorld-Verified;发布首表报告值。
条件
First-attempt success rate at 1080p, maximum 100 action steps, average over five runs. System Card table/prose says 78.0%; Figure 8.9.4.A labels 77.9%.

关联材料,仅供追溯

First-attempt success rate77.9%原文 77.9% · 单位 %
公布 未确认

结果冲突待核对,精确排名不采用。

发布表报告值
设置
System Card §8.1 standard configuration: adaptive thinking at max effort; default temperature/top_p; averaged over 5 trials. Context sizes are evaluation-dependent and do not exceed 1M tokens; retain all section-specific exceptions on their results.
未知项
Per-benchmark tools, context/token cap, resolution, trial count, harness and other exceptions are recorded on each result when stated.
工具
工具条件未记录
推理预算
max
框架
未记录
协议
OSWorld-Verified; first-attempt task success; updated agent scaffolding.
条件
Figure 8.9.4.A label. Table 8.1.A and §8.9.4 prose report 78.0%; this candidate records the printed chart value and must remain conflict-marked. 1080p; 100-step cap; five runs.

直接支持该成绩

在排名中查看此成绩