Claude Opus 4.6 · BioPipelineBench(Anthropic internal unpublished computational-biology evaluation; version and task count unspecified) 53.1% · 附件证据(ev-op46-card-biopipeline)

Claude Opus 4.6 的 BioPipelineBench · Anthropic internal unpublished computational-biology evaluation; version and task count unspecified 官方出处。Life science internal · no extended thinking · 53.1%。见《Claude Opus 4.6 System Card》PDF pp.45–46, §2.22 Computational Biology, BioPipelineBench paragraph and Figure 2.22.A。证据 ev-op46-card-biopipeline。

原文与定位

Claude Opus 4.6 · BioPipelineBench(Anthropic internal unpublished computational-biology evaluation; version and task count unspecified),Life science internal · no extended thinking配置,Score 53.1%(具体成绩公布时间未确认)。见《Claude Opus 4.6 System Card》PDF pp.45–46, §2.22 Computational Biology, BioPipelineBench paragraph and Figure 2.22.A。 查看成绩条件

BioPipelineBench score printed as ‘53.1.%’ (punctuation typo); interpreted as 53.1%. This is an unpublished Anthropic internal capability evaluation; all life-science tasks had bash/package managers and no extended thinking.
官方材料
Claude Opus 4.6 System Card · PDF 第 45 页
定位
附件 · PDF pp.45–46, §2.22 Computational Biology, BioPipelineBench paragraph and Figure 2.22.A
材料版本
Current PDF/TXT; local source 本地留存副本 and 本地留存副本; 213 pages; changelog through 2026-03-06; full text and visuals reviewed.
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Claude Opus 4.6 · BioPipelineBench(Anthropic internal unpublished computational-biology evaluation; version and task count unspecified),Life science internal · no extended thinking配置,Score 53.1%(具体成绩公布时间未确认)。官方出处 Claude Opus 4.6 System Card(附件,Current PDF/TXT; local source 本地留存副本 and 本地留存副本; 213 pages; changelog through 2026-03-06; full text and visuals reviewed.)PDF pp.45–46, §2.22 Computational Biology, BioPipelineBench paragraph and Figure 2.22.A,https://www.anthropic.com/claude-opus-4-6-system-card#page=45。Benchlog https://benchlog.lekshome.cc/evidence/claude-opus-4-6/ev-op46-card-biopipeline/

关联成绩与条件

Claude Opus 4.6 · BioPipelineBench 成绩与配置
指标与成绩配置与评测条件证据关系
Score53.1%原文 53.1.% · 单位 %
公布 未确认
Life science internal · no extended thinking
设置
Anthropic 内部生命科学评测;bash 与 package managers;without extended thinking。
未知项
各项为非公开内部评测;trial 数与 effort 未说明。
工具
允许工具
推理预算
未记录
框架
未记录
协议
BioPipelineBench;computational biology。
条件
Unpublished Anthropic internal capability evaluation; bash/package managers available; without extended thinking; version, effort and trial count not disclosed. Source prints 53.1.% (punctuation typo); numeric value is recorded as 53.1%.

直接支持该成绩

在排名中查看此成绩