Artificial Analysis Intelligence Index v5 Widens Open-vs-Closed Model Gap to 12 Points

Artificial Analysis released Intelligence Index v5 on Oct 27, with 60% of the suite now backed by private test sets and median output rising from 35K to 59K tokens per task. Claude Opus 5.5 tops at 58, GPT-6 Astra at 57, GPT-6.1 Sol at 55, Gemini 4 Argon at 50; the leading open-weight (Kimi K3.1 Pro, weights pending) sits at 46, widening the open-vs-closed gap to 12 points from 3 in v4. Released open weights DeepSeek V4.1 Flash and GLM-5.3 both score 32, 26 points behind the leader.