Independent tests put Kimi K3.1 Pro fourth of 215 models, but slow and verbose
Artificial Analysis scores Moonshot's new model 55 on its Intelligence Index, fourth of 215 and on the price-performance frontier at $2/$10 per million tokens. The catch: it burned 150 million output tokens on the suite, against an 87 million median, and streams at 73.5 tokens a second. Weights are still pending.
Artificial Analysis




