Claude Opus 5.5 Takes Top Spot as Artificial Analysis Launches Intelligence Index v5

Artificial Analysis rebuilt its widely-cited AI benchmark this week, putting 60% of the score behind private test sets that labs can't study in advance. The reshuffle put Claude Opus 5.5 in first place and left open-weight models further behind than before. Artificial Analysis launched Intelligence Index v5 in late October, and the leaderboard moved. Claude Opus 5.5, running at max reasoning effort, now sits at the top with a score of 58, ahead of OpenAI's GPT-6 Astra at 57 and GPT-6.1 Sol at 55. The three labs are separated by two points combined, which tells you how tight the frontier has become even as the test itself got harder. The bigger story is what changed underneath the scores, not just who's on top.

First reported by Startup Fortune. This page is a short excerpt; the full text is theirs.