Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations) · 附件证据(ev55-visual-healthcare-effort)

Claude Sonnet 5.5 的 HealthBench · 48,000+ rubric items across 5,000 multi-turn conversations 官方出处。见《Claude Sonnet 5.5 System Card》§8.15, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription。证据 ev55-visual-healthcare-effort。

原文与定位

Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Low effort配置,Length-adjusted score 64.9%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.15, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Medium effort配置,Length-adjusted score 64.9%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.15, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),High effort配置,Length-adjusted score 64.9%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.15, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription。 查看成绩条件

Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Xhigh effort配置,Length-adjusted score 64.7%(具体成绩公布时间未确认)。见《Claude Sonnet 5.5 System Card》§8.15, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription。 查看成绩条件

HealthBench labels: low to high 64.9%; xhigh 64.7%; max 65.4%. HealthBench Professional labels: low 67.9%; medium 67.6%; high 67.8%; xhigh 68.2%; max 69.2%. PhysicianBench labels: low 27.2%; medium 30.0%; high 47.6%; xhigh 56.4%; max 63.2%.
官方材料
Claude Sonnet 5.5 System Card · PDF 第 138 页
定位
附件 · §8.15, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription
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),Low effort配置,Length-adjusted score 64.9%(具体成绩公布时间未确认);Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Medium effort配置,Length-adjusted score 64.9%(具体成绩公布时间未确认);Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),High effort配置,Length-adjusted score 64.9%(具体成绩公布时间未确认);Claude Sonnet 5.5 · HealthBench(48,000+ rubric items across 5,000 multi-turn conversations),Xhigh effort配置,Length-adjusted score 64.7%(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, Figure 8.15.B, PDF p.138; Claude Sonnet 5.5 chart-label transcription,https://www.anthropic.com/claude-sonnet-5-5-system-card#page=138。Benchlog https://benchlog.lekshome.cc/evidence/claude-sonnet-5-5/ev55-visual-healthcare-effort/

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
Length-adjusted score64.9%原文 64.9% · 单位 %
公布 未确认
Low effort
设置
Low reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
工具条件未记录
推理预算
low
框架
未记录
协议
HealthBench length-adjusted score.
条件
Figure 8.15.B labels the shared Low-to-High result as 64.9%; this row uses the Low effort config. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench responses.

直接支持该成绩

在排名中查看此成绩
Length-adjusted score64.9%原文 64.9% · 单位 %
公布 未确认
Medium effort
设置
Medium reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
工具条件未记录
推理预算
medium
框架
未记录
协议
HealthBench length-adjusted score.
条件
Figure 8.15.B labels the shared Low-to-High result as 64.9%; this row uses the Medium effort config. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench responses.

直接支持该成绩

在排名中查看此成绩
Length-adjusted score64.9%原文 64.9% · 单位 %
公布 未确认
High effort
设置
High reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
工具条件未记录
推理预算
high
框架
未记录
协议
HealthBench length-adjusted score.
条件
Figure 8.15.B labels the shared Low-to-High result as 64.9%; this row uses the High effort config. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench responses.

直接支持该成绩

在排名中查看此成绩
Length-adjusted score64.7%原文 64.7% · 单位 %
公布 未确认
Xhigh effort
设置
Xhigh reasoning effort.
未知项
Sampling settings not stated unless recorded per result.
工具
工具条件未记录
推理预算
xhigh
框架
未记录
协议
HealthBench length-adjusted score.
条件
Figure labels the effort Xhigh. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench responses.

直接支持该成绩

在排名中查看此成绩