@imranpatel_si Yes sir, I understand it's an aggregate but the rankings don't really make sense. Even apart from the regression that this bench shows, it also shows that Sonnet is only two points behind Opus. and better then 6.1 Sol Which isn't true
A REGRESSION FOR CODING! Astra 6.2 & Sol 6.2.
Why OPENAI why!?!?!?!?!
GPT-6.2 Astra is flat with GPT-6 Astra, and GPT-6.2 Sol loses 3 points relative to GPT-6.1 Sol in the Artificial Analysis Coding Agent Index (Codex harness at max effort).