AnalysisAI ModelsOctober 28, 2026

Artificial Analysis launches Intelligence Index v5 with private benchmarks

Read original source →artificialanalysis.ai

60% of the v5 Index now rests on private test sets or held-out answers, with 2 new benchmarks (Cognition's FrontierCode 2.0, Terminal-Bench-Science) and 3 upgrades. Claude Opus 5.5 (max) leads at 58, ahead of GPT-6 Astra (max, 57) and GPT-6.1 Sol.

How this story unfolded

2 days · 1 report · 11 community posts · 12 of 13 shown

  1. Oct 27
  2. Oct 28
  3. Oct 29

More stories today

Open the live feed