Claude Sonnet 5.5 · HealthBench Professional · 525 physician-authored conversations · 成绩与配置

Claude Sonnet 5.5 的 HealthBench Professional · 525 physician-authored conversations 记录。本次发布有 6 条成绩记录,含 6 条精确值(可含历史记录),附 2 条来源定位。

成绩与评测条件

Claude Sonnet 5.5 · HealthBench Professional 成绩与配置
指标与成绩配置与评测条件出处
Raw score77.1%原文 77.1% · 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 Professional, max effort.
条件
Raw, unadjusted score.
在排名中查看此成绩
Length-adjusted score69.2%原文 69.2% length-adjusted · 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 Professional, max effort.
条件
Length adjustment uses the benchmark developer's method; distinct from raw score.
在排名中查看此成绩
Length-adjusted score67.9%原文 67.9% · 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 Professional length-adjusted score.
条件
Figure labels the effort Low. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench Professional responses.
在排名中查看此成绩
Length-adjusted score67.6%原文 67.6% · 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 Professional length-adjusted score.
条件
Figure labels the effort Medium. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench Professional responses.
在排名中查看此成绩
Length-adjusted score67.8%原文 67.8% · 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 Professional length-adjusted score.
条件
Figure labels the effort High. Solid scores reflect length adjustment; all models used safety classifiers and Claude Opus 4.8 graded HealthBench Professional responses.
在排名中查看此成绩
Length-adjusted score68.2%原文 68.2% · 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 Professional 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 Professional responses.
在排名中查看此成绩

出处与披露

测试版本与指标

别名 HealthBench Professional

  • Raw score · % · 越高越好
  • Length-adjusted score · % · 越高越好
  • Score · % · 越高越好
  • Mean response length · characters · 排序方向未确认
  • 长度校正分(0–100) · score · 越高越好
  • 未校正分(0–100) · score · 越高越好
  • 相对 GPT-5.4 的长度校正分差 · points · 排序方向未确认
  • Unadjusted score · % · 越高越好