New AA intelligence index v5 with “Private tests” conveniently upranks GPT, Google’s models, while open weights chinese models downrank

Overview

The new Artificial Analysis intelligence index v5, is being criticised, because its introduction of "trust me bro" private tests on private datasets, significantly downranking current open weights models on the index. Just after its published, google announces Carbon 4, which is placed on the top of the index. Kimi 3.1 Pro, Deepseek 4.1 Flash,Xiaomi mimo v2.6 Pro and other open models drop significantly in intelligence points in this new index. I think its fairly suspicious, what do yo

SpinGraph analysis pending — check back after processing.