Claude Sonnet 5.5 · SpatialBench Verified(Verified) 72.5% · 附件证据(ev55-spatialbench)

Claude Sonnet 5.5 的 SpatialBench Verified · Verified 官方出处。Adaptive max with biology safeguards disabled · 72.5%。见《Claude Sonnet 5.5 System Card》§8.17.2, PDF p.143。证据 ev55-spatialbench。

原文与定位

Claude Sonnet 5.5 · SpatialBench Verified(Verified),Adaptive max with biology safeguards disabled配置,Score 72.5%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.17.2, PDF p.143。 查看成绩条件

Claude Sonnet 5.5 performs on par with Claude Sonnet 5 on both (SpatialBench Verified 72.5% vs.
官方材料
Claude Sonnet 5.5 System Card · PDF 第 143 页
定位
附件 · §8.17.2, PDF p.143
材料版本
PDF cover dated 2026-09-28; 148 pages; collected 2026-11-10
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 Sonnet 5.5 · SpatialBench Verified(Verified),Adaptive max with biology safeguards disabled配置,Score 72.5%(具体成绩公布时间未确认)。官方出处 Claude Sonnet 5.5 System Card(附件,PDF cover dated 2026-09-28; 148 pages; collected 2026-11-10)§8.17.2, PDF p.143,https://www.anthropic.com/claude-sonnet-5-5-system-card#page=143。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-5-5/ev55-spatialbench/

关联成绩与条件

Claude Sonnet 5.5 · SpatialBench Verified 成绩与配置
指标与成绩配置与评测条件证据关系
Score72.5%原文 72.5% · 单位 %
公布 未确认
Adaptive max with biology safeguards disabled
设置
Adaptive thinking at max effort; API evaluation with biology safeguards turned off.
未知项
Sampling settings follow each benchmark's report.
工具
允许工具
推理预算
max
框架
未记录
协议
SpatialBench Verified over 115 externally validated problems.
条件
Bash, file editor and prescribed packages available; API evaluation with biology safeguards disabled; adaptive max, five attempts per problem.

直接支持该成绩