Confirmed
Artificial Analysis releases Intelligence Index v5 with wider open-weights gap
Kimi K3.1 Pro leads at 46 but its weights are not yet released. The top released open-weights models, DeepSeek V4.1 Flash 1017 and GLM-5.3, score 32, and Claude Haiku 5.5 scores 36. Sixty percent of the index relies on private test sets or held-out answers, so results cannot be reproduced.
Every outlet that ran it 1 article
r/LocalLLaMACommunityLead