Claude Fable 5.2 · HealthBench Professional(525 physician-authored conversations) · 附件证据(ev-f52-healthbench-pro)
Claude Fable 5.2 的 HealthBench Professional · 525 physician-authored conversations 官方出处。见《Claude Fable 5.2 and Claude Mythos 5.2 System Card》PDF p.246, Figure 8.16.2.A, HealthBench Professional(已渲染并视觉核读数值标签)。证据 ev-f52-healthbench-pro。
原文与定位
Claude Fable 5.2 · HealthBench Professional(525 physician-authored conversations),Adaptive Max配置,Raw score 71.1%(具体成绩公布时间未确认)。见《Claude Fable 5.2 and Claude Mythos 5.2 System Card》PDF p.246, Figure 8.16.2.A, HealthBench Professional(已渲染并视觉核读数值标签)。 查看成绩条件
Claude Fable 5.2 · HealthBench Professional(525 physician-authored conversations),Adaptive Max配置,Length-adjusted score 67.7%(具体成绩公布时间未确认)。见《Claude Fable 5.2 and Claude Mythos 5.2 System Card》PDF p.246, Figure 8.16.2.A, HealthBench Professional(已渲染并视觉核读数值标签)。 查看成绩条件
Artificial Analysis scores graded by GPT-5.4 at low reasoning effort: Fable 5.2 raw 71.1% and length-adjusted 67.7% at max effort.
- 官方材料
- Claude Fable 5.2 and Claude Mythos 5.2 System Card · PDF 第 246 页
- 定位
- 附件 · PDF p.246, Figure 8.16.2.A, HealthBench Professional(已渲染并视觉核读数值标签)
- 材料版本
- Official System Card; 264-page PDF, cover dated November 9, 2026; current PDF/TXT saved as 本地留存副本 and .txt; PDF pp.1-264 text read page-by-page; all 80 image/vector-figure pages in pp.1-206 were visually checked in 21 sheets, and capability figures in pp.207-254 were visually checked; no unread pages or figures remain.
- 版本公布
- 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 Fable 5.2 · HealthBench Professional(525 physician-authored conversations),Adaptive Max配置,Raw score 71.1%(具体成绩公布时间未确认);Claude Fable 5.2 · HealthBench Professional(525 physician-authored conversations),Adaptive Max配置,Length-adjusted score 67.7%(具体成绩公布时间未确认)。官方出处 Claude Fable 5.2 and Claude Mythos 5.2 System Card(附件,Official System Card; 264-page PDF, cover dated November 9, 2026; current PDF/TXT saved as 本地留存副本 and .txt; PDF pp.1-264 text read page-by-page; all 80 image/vector-figure pages in pp.1-206 were visually checked in 21 sheets, and capability figures in pp.207-254 were visually checked; no unread pages or figures remain.)PDF p.246, Figure 8.16.2.A, HealthBench Professional(已渲染并视觉核读数值标签),https://www.anthropic.com/document/claude-fable-5-2-mythos-5-2-system-card#page=246。Benchlog https://benchlog.lekshome.cc/evidence/claude-fable-and-mythos-5-2/ev-f52-healthbench-pro/
关联成绩与条件
| 指标与成绩 | 配置与评测条件 | 证据关系 |
|---|---|---|
| Raw score71.1%原文 71.1% · 单位 % 公布 未确认 | Adaptive Max
| 直接支持该成绩 在排名中查看此成绩 |
| Length-adjusted score67.7%原文 67.7% · 单位 % 公布 未确认 | Adaptive Max
| 直接支持该成绩 在排名中查看此成绩 |