Claude Opus 4.7 · LAB-Bench FigQA(FutureHouse FigQA; prompting and tools are recorded per result) · 附件证据(ev-op47-card-labbench-resolution-ablation)

Claude Opus 4.7 的 LAB-Bench FigQA · FutureHouse FigQA; prompting and tools are recorded per result 官方出处。见《Claude Opus 4.7 System Card》§8.9.1 and Figure 8.9.1.B, PDF pp.203-204。证据 ev-op47-card-labbench-resolution-ablation。

原文与定位

Claude Opus 4.7 · LAB-Bench FigQA(FutureHouse FigQA; prompting and tools are recorded per result),LAB-Bench FigQA,无工具,旧分辨率配置,Score 74%(具体成绩公布时间未确认)。见《Claude Opus 4.7 System Card》§8.9.1 and Figure 8.9.1.B, PDF pp.203-204。 查看成绩条件

Claude Opus 4.7 · LAB-Bench FigQA(FutureHouse FigQA; prompting and tools are recorded per result),LAB-Bench FigQA,Python工具,旧分辨率配置,Score 85.8%(具体成绩公布时间未确认)。见《Claude Opus 4.7 System Card》§8.9.1 and Figure 8.9.1.B, PDF pp.203-204。 查看成绩条件

At the old 1568px single-axis / 1.15MP maximum image resolution, adaptive thinking at max effort yields 74.0% without tools and 85.8% with Python tools. Scores are averaged over five runs.
官方材料
Claude Opus 4.7 System Card · PDF 第 203 页
定位
附件 · §8.9.1 and Figure 8.9.1.B, PDF pp.203-204
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 · LAB-Bench FigQA(FutureHouse FigQA; prompting and tools are recorded per result),LAB-Bench FigQA,无工具,旧分辨率配置,Score 74%(具体成绩公布时间未确认);Claude Opus 4.7 · LAB-Bench FigQA(FutureHouse FigQA; prompting and tools are recorded per result),LAB-Bench FigQA,Python工具,旧分辨率配置,Score 85.8%(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.)§8.9.1 and Figure 8.9.1.B, PDF pp.203-204,https://www.anthropic.com/claude-opus-4-7-system-card#page=203。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-4-7/ev-op47-card-labbench-resolution-ablation/

关联成绩与条件

Claude Opus 4.7 · LAB-Bench FigQA 成绩与配置
指标与成绩配置与评测条件证据关系
Score74%原文 74.0% · 单位 %
公布 未确认
LAB-Bench FigQA,无工具,旧分辨率
设置
Adaptive thinking, max effort; no tools; 5-run average; old image limit (1568px single dimension, 1.15MP total).
未知项
未另记
工具
禁止工具
推理预算
max
框架
未记录
协议
LAB-Bench FigQA; no tools; adaptive thinking; max effort; old image resolution.
条件
Five-run average; image maximum 1568px on one dimension and 1.15MP total. Separate resolution ablation, not the current-resolution 78.6% result.

直接支持该成绩

在排名中查看此成绩
Score85.8%原文 85.8% · 单位 %
公布 未确认
LAB-Bench FigQA,Python工具,旧分辨率
设置
Adaptive thinking, max effort; Python tools; 5-run average; old image limit (1568px single dimension, 1.15MP total).
未知项
未另记
工具
允许工具
推理预算
max
框架
未记录
协议
LAB-Bench FigQA; Python tools; adaptive thinking; max effort; old image resolution.
条件
Five-run average; image maximum 1568px on one dimension and 1.15MP total. Separate resolution ablation, not the current-resolution 86.4% result.

直接支持该成绩

在排名中查看此成绩