GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified) · 附件证据(ev-gpt6-luna-hbh)

GPT-6 Luna 的 HealthBench Hard · System-card evaluation; exact dataset version unspecified 官方出处。见《Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (PDF)》PDF p.11 / Table 7 / HealthBench Hard row / GPT-6 Luna column。证据 ev-gpt6-luna-hbh。

原文与定位

GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified),GPT-6 Luna, system-card comparison configuration unspecified配置,Length-adjusted score (0–100) 31.4 score(具体成绩公布时间未确认)。见《Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (PDF)》PDF p.11 / Table 7 / HealthBench Hard row / GPT-6 Luna column。 查看成绩条件

GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified),GPT-6 Luna, system-card comparison configuration unspecified配置,Unadjusted score (0–100) 25.4 score(具体成绩公布时间未确认)。见《Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (PDF)》PDF p.11 / Table 7 / HealthBench Hard row / GPT-6 Luna column。 查看成绩条件

GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified),GPT-6 Luna, system-card comparison configuration unspecified配置,Mean response length 1241 characters(具体成绩公布时间未确认)。见《Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (PDF)》PDF p.11 / Table 7 / HealthBench Hard row / GPT-6 Luna column。 查看成绩条件

GPT-6 Luna: 31.4 (25.4, 1241).
官方材料
Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (PDF) · PDF 第 11 页
定位
附件 · PDF p.11 / Table 7 / HealthBench Hard row / GPT-6 Luna column
材料版本
47-page PDF; full text extracted, Table 7 read for HealthBench values, remaining sections not fully reviewed.
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

GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified),GPT-6 Luna, system-card comparison configuration unspecified配置,Length-adjusted score (0–100) 31.4 score(具体成绩公布时间未确认);GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified),GPT-6 Luna, system-card comparison configuration unspecified配置,Unadjusted score (0–100) 25.4 score(具体成绩公布时间未确认);GPT-6 Luna · HealthBench Hard(System-card evaluation; exact dataset version unspecified),GPT-6 Luna, system-card comparison configuration unspecified配置,Mean response length 1241 characters(具体成绩公布时间未确认)。官方出处 Addendum to GPT-6 Astra System Card: GPT-6.1 Sol (PDF)(附件,47-page PDF; full text extracted, Table 7 read for HealthBench values, remaining sections not fully reviewed.)PDF p.11 / Table 7 / HealthBench Hard row / GPT-6 Luna column,https://deploymentsafety.openai.com/gpt-6-1-sol/gpt-6-1-sol.pdf#page=11。Benchlog https://benchlog.lekshome.cc/evidence/gpt-6-luna/ev-gpt6-luna-hbh/

关联成绩与条件

GPT-6 Luna · HealthBench Hard 成绩与配置
指标与成绩配置与评测条件证据关系
Length-adjusted score (0–100)31.4 score原文 31.4 · 单位 score
公布 未确认
GPT-6 Luna, system-card comparison configuration unspecified
设置
GPT-6 Luna; 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 Luna; effort unspecified. The card warns comparison values for previously launched models may reflect later versions and differ from launch.

直接支持该成绩

在排名中查看此成绩
Unadjusted score (0–100)25.4 score原文 25.4 · 单位 score
公布 未确认
GPT-6 Luna, system-card comparison configuration unspecified
设置
GPT-6 Luna; 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 Luna; effort unspecified. The card warns comparison values for previously launched models may reflect later versions and differ from launch.

直接支持该成绩

在排名中查看此成绩
Mean response length1241 characters原文 1241 · 单位 characters
公布 未确认
GPT-6 Luna, system-card comparison configuration unspecified
设置
GPT-6 Luna; 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 Luna; effort unspecified. The card warns comparison values for previously launched models may reflect later versions and differ from launch.

直接支持该成绩

在排名中查看此成绩