Kimi K3.1 Pro debuted with a score of 55 on the Artificial Analysis Intelligence Index, ranking fourth overall and improving by 11 points compared with Kimi K3.

The model achieved the highest result of any tested model on Terminal-Bench 4.0, scoring 67%, and showed major improvements in agentic workloads. Artificial Analysis said K3.1 Pro would become the highest-scoring open-weight model on its index once Moonshot publishes the weights.

Source: artificialanalysis.ai