AI Briefing
KOSign in

Artificial Analysis Intelligence Index v5 widens open-weight gap to 12 points

·2026.10.28 12:36

Key point

The update makes 60% of the index based on private test sets, preventing independent reproduction of results.

Details

  • Artificial Analysis Intelligence Index updated to v5, resulting in a wider gap between open-weight and closed models.
  • The open-weights gap widens from 3 points to 12 points.
  • Kimi K3.1 Pro (weights not yet released) is the top open-weights model at 46.
  • Among released weights, DeepSeek V4.1 Flash 1017 (max) and GLM-5.3 (max) lead at 32, which is 26 points behind the leader.
  • These released open-weight models are below Claude Haiku 5.5 (max, 36).
  • 60% of the index is based on evaluations with private test sets or held-out answers, making it impossible to reproduce the results.
  • Some users question the accuracy based on their experiences with models, citing GPT-Terra-6.1 > Fable-5.1.
artificial-analysisllm-benchmarkopen-weight-modelsevaluation-metricsmodel-ranking
Read originalShare on Threads

This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.

Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.

Related

RedditMeta's 816B Muse Spark Open scores 39 on Artificial Analysis, trailing 180B Qwen3.8-Flash-NextGeekNewsQwen3.8 27B Records 52 Points on Artificial Analysis Intelligence IndexRedditArtificial Analysis Releases Intelligence Index v4.3Hacker NewsArtificial Analysis Releases Small LLM Inference Benchmark for Mobile DevicesHacker NewsQwen3.8 27B Scores 52 on Artificial AnalysisGeekNewsQwen3.8 Max Ranks First Overall on Agentic Index