Event ·Multi-source

Artificial Analysis 智能指数 v5 上线：60% 权重改用私有测试集

Artificial Analysis 发布 Intelligence Index v5，将 60% 的评分权重交给使用私有测试集或保留答案的评测，以降低基准污染。榜单上 Claude Opus 5.5 以微弱优势领先 GPT-6 Astra 和 GPT-6.1 Sol。分项来看，GPT-6 Astra 在科学类基准上略占上风，Claude Opus 5.5 在编程任务中领先，GPT-6.1 Sol 则在性价比上表现最好。

Current read

Artificial Analysis 把 60% 评测改成私有题库防作弊，现在 Claude Opus 5.5 排第一，选模型前可以看看这份新榜。

11 sources41° AI heatLast updated 2026/10/28 02:56:11

Evidence

11 sources

Primary sourcekimmonismus

2026/10/28 02:56:11View source

Related source 1Artificial Analysis

2026/10/28 02:11:22View source

Related source 2TestingCatalog

2026/10/26 18:23:18View source

Related source 3The Rundown AI

2026/10/26 18:32:26View source

Related source 4IT之家

2026/10/26 23:26:19View source

Related source 5Ethan Mollick

2026/10/27 13:01:04View source

Related source 6arXiv: DeepSeek

2026/10/26 05:58:10View source

Related source 7Shubham Saboo

2026/10/27 03:01:36View source

Related source 8量子位

2026/10/27 08:27:46View source

Related source 9AIGCLINK

2026/10/26 23:38:03View source

Related source 10Simon Willison’s Weblog

2026/10/26 23:53:07View source

Conflict check

Dedup data only shows that these sources discuss the same event, not that they agree. There is no structured supporting or refuting evidence yet, so no conflict is inferred automatically.

Open the story