GPT-6 Sol · HealthBench · 48,000+ rubric items across 5,000 multi-turn conversations · 成绩与配置

GPT-6 Sol 的 HealthBench · 48,000+ rubric items across 5,000 multi-turn conversations 记录。本次发布有 3 条成绩记录,含 3 条精确值(可含历史记录),附 1 条来源定位。

成绩与评测条件

GPT-6 Sol · HealthBench 成绩与配置
指标与成绩配置与评测条件出处
长度校正分(0–100)53.2 score原文 53.2 · 单位 score
公布 未确认
GPT-6 Sol, system-card comparison configuration unspecified
设置
GPT-6 Sol; system-card comparison table does not identify reasoning effort or serving configuration.
未知项
Reasoning effort not stated、Sampling and tool settings not stated
工具
工具条件未记录
推理预算
未记录
框架
未记录
协议
System-card Table 7: length-adjusted score (unadjusted score, mean response length in characters).
条件
GPT-6 Sol; effort unspecified. The card warns comparison values for previously launched models may reflect later versions and differ from launch.
在排名中查看此成绩
未校正分(0–100)47.1 score原文 47.1 · 单位 score
公布 未确认
GPT-6 Sol, system-card comparison configuration unspecified
设置
GPT-6 Sol; system-card comparison table does not identify reasoning effort or serving configuration.
未知项
Reasoning effort not stated、Sampling and tool settings not stated
工具
工具条件未记录
推理预算
未记录
框架
未记录
协议
System-card Table 7: unadjusted score paired with the length-adjusted score.
条件
GPT-6 Sol; effort unspecified. The card warns comparison values for previously launched models may reflect later versions and differ from launch.
在排名中查看此成绩
Mean response length977 characters原文 977 · 单位 characters
公布 未确认
GPT-6 Sol, system-card comparison configuration unspecified
设置
GPT-6 Sol; system-card comparison table does not identify reasoning effort or serving configuration.
未知项
Reasoning effort not stated、Sampling and tool settings not stated
工具
工具条件未记录
推理预算
未记录
框架
未记录
协议
System-card Table 7: mean response length paired with adjusted and unadjusted scores.
条件
GPT-6 Sol; effort unspecified. The card warns comparison values for previously launched models may reflect later versions and differ from launch.
在排名中查看此成绩

出处与披露

测试版本与指标

别名 HealthBench

  • Length-adjusted score · % · 越高越好
  • Raw 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