Claude Opus 5 · Legal Agent Benchmark(Public full set) · 附件证据(ev-op5-card-lab-full)

Claude Opus 5 的 Legal Agent Benchmark · Public full set 官方出处。见《Claude Opus 5 System Card》System Card PDF p.181, §8.13.3 Legal Agent Benchmark。证据 ev-op5-card-lab-full。

原文与定位

Claude Opus 5 · Legal Agent Benchmark(Public full set),Legal Agent Benchmark · full set配置,All-pass rate 23.58%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.181, §8.13.3 Legal Agent Benchmark。 查看成绩条件

Claude Opus 5 · Legal Agent Benchmark(Public full set),Legal Agent Benchmark · full set配置,Mean criterion-pass rate 93.74%(具体成绩公布时间未确认)。见《Claude Opus 5 System Card》System Card PDF p.181, §8.13.3 Legal Agent Benchmark。 查看成绩条件

On 1,235 evaluated tasks after 16 data-defect exclusions, Opus 5 achieved 23.58% (±0.48, n=5) all-pass and 93.74% mean criterion-pass rate at adaptive max effort. Public Messages API production safeguards/fallback were active; Anthropic used an internal reimplementation with bash and Python tools and Sonnet 4.6 as judge.
官方材料
Claude Opus 5 System Card · PDF 第 181 页
定位
附件 · System Card PDF p.181, §8.13.3 Legal Agent Benchmark
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 · Legal Agent Benchmark(Public full set),Legal Agent Benchmark · full set配置,All-pass rate 23.58%(具体成绩公布时间未确认);Claude Opus 5 · Legal Agent Benchmark(Public full set),Legal Agent Benchmark · full set配置,Mean criterion-pass rate 93.74%(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 p.181, §8.13.3 Legal Agent Benchmark,https://www.anthropic.com/claude-opus-5-system-card#page=181。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-5/ev-op5-card-lab-full/

关联成绩与条件

Claude Opus 5 · Legal Agent Benchmark 成绩与配置
指标与成绩配置与评测条件证据关系
All-pass rate23.58%原文 23.58% ±0.48 (n=5) · 单位 %
公布 未确认
Legal Agent Benchmark · full set
设置
1,235 evaluated tasks after 16 data-defect exclusions; adaptive thinking at max effort; public Messages API with production safeguards and fallback to Opus 4.8; internal reimplementation with bash and Python tools; five-run all-pass average; Sonnet 4.6 judge.
未知项
未另记
工具
允许工具
推理预算
max
框架
未记录
协议
Legal Agent Benchmark; full Anthropic evaluation set; all-pass rate.
条件
1,235 tasks after 16 data-defect exclusions; adaptive max effort; public Messages API safeguards/fallback; internal reduced-tool reimplementation; five-run average.

直接支持该成绩

在排名中查看此成绩
Mean criterion-pass rate93.74%原文 93.74% · 单位 %
公布 未确认
Legal Agent Benchmark · full set
设置
1,235 evaluated tasks after 16 data-defect exclusions; adaptive thinking at max effort; public Messages API with production safeguards and fallback to Opus 4.8; internal reimplementation with bash and Python tools; five-run all-pass average; Sonnet 4.6 judge.
未知项
未另记
工具
允许工具
推理预算
max
框架
未记录
协议
Legal Agent Benchmark; full Anthropic evaluation set; mean criterion-pass rate.
条件
1,235 tasks after 16 data-defect exclusions; adaptive max effort; public Messages API safeguards/fallback; internal reduced-tool reimplementation; five-run evaluation.

直接支持该成绩

在排名中查看此成绩