AITOPAI热报
ModelsMulti-source

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

kimmonismus@kimmonismus
Post· 11 sources
View the post
Why we picked it

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。

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。

Original · kimmonismus

Artificial Analysis’ Intelligence Index v5 now bases 60% of its score on evaluations with private test sets or held-out answers, with Claude Opus 5.5 narrowly leading GPT-6 Astra and GPT-6.1 Sol.

As expected, GPT-6-Astra currently appears to be slightly ahead in scientific benchmarks, while Opus 5.5 leads in coding. GPT-6.1-Sol offers unparalleled price-performance.

Good to see we are getting more private tests.

  • Artificial Analysis02:11原文
  • TestingCatalog10-26 18:23原文
  • The Rundown AI10-26 18:32原文
  • IT之家10-26 23:26原文
  • Ethan Mollick10-27 13:01原文
  • arXiv: DeepSeek10-26 05:58原文
  • Shubham Saboo10-27 03:01原文
  • 量子位10-27 08:27原文
  • AIGCLINK10-26 23:38原文
  • Simon Willison’s Weblog10-26 23:53原文
#Artificial Analysis#Claude Opus 5.5#GPT-6#基准测试#推理模型

Keep reading

Recent stories with the same tag or section
ModelsUS vs ChinaMulti-source74°10:17
Chinese contentArtificial Analysis 发布 Intelligence Index v5，新增两个基准并引入私有测试集

模型评测榜 v5 版来了，60% 改用私有题防刷榜，Claude Opus 5.5 拿了第一，开源和闭源差距拉大到 12 分，选模型前可以看看。

#Artificial Analysis#Claude Opus 5.5#GPT-6#Kimi K3.1 Pro
Event
官方账号Artificial Analysis@ArtificialAnlys11 sourcesSource
ModelsUS vs ChinaMulti-sourceSelected01:57
CritPt v2 Benchmark Launches with Agentic Code Execution

The 54-point gap between GPT-5 and GPT-6 Astra in about a year shows how quickly frontier models are absorbing research-level physics when given code execution. Tracking these scores before the late October Intelligence Index update helps researchers and benchmark watchers pick models for scientific workflows. Compare against Claude Opus 5.5 at 78% when choosing a frontier model for agentic computation tasks.

#GPT-6#Claude Opus 5.5#CritPt#Artificial Analysis
Event
官方账号Artificial Analysis@ArtificialAnlys11 sourcesSource