Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations) 69.4% · 附件证据(ev55-healthbench)

Claude Sonnet 5.5 的 HealthBench · 48,000+ rubric items across 5,000 multi-turn conversations 官方出处。Max effort · 69.4%。见《Claude Sonnet 5.5 System Card》§8.15.1 and Figures 8.15.A-B, PDF p.138。证据 ev55-healthbench。

原文与定位

Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Max effort配置,Raw score 69.4%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.15.1 and Figures 8.15.A-B, PDF p.138。 查看成绩条件

Before length adjustment it is level with Claude Opus 5.5 from low to xhigh effort and slightly ahead at max (69.4% against 68.1%).
官方材料
Claude Sonnet 5.5 System Card · PDF 第 138 页
定位
附件 · §8.15.1 and Figures 8.15.A-B, PDF p.138
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 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Max effort配置,Raw score 69.4%(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)§8.15.1 and Figures 8.15.A-B, PDF p.138,https://www.anthropic.com/claude-sonnet-5-5-system-card#page=138。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-5-5/ev55-healthbench/

关联成绩与条件

Claude Sonnet 5.5 · HealthBench 成绩与配置
指标与成绩配置与评测条件证据关系
Raw score69.4%原文 69.4% · 单位 %
公布 未确认
Max effort
设置
Max reasoning effort.
未知项
Adaptive thinking status is benchmark-specific.
工具
工具条件未记录
推理预算
max
框架
未记录
协议
HealthBench, max effort.
条件
Raw, unadjusted score. The report gives the length-adjusted score over all effort levels only as a 64.7%-65.4% range; per-effort exact chart values were not recovered.

直接支持该成绩

在排名中查看此成绩