Kimi K3.1 Pro Tested on CrusaderBench, Trails Opus 5.5

Kimi K3.1 Pro scored 43.3 on CrusaderBench, an independent benchmark where AI models manage a medieval village in Stronghold Crusader via screenshots and keyboard-mouse control, slightly behind Claude Opus 5.5 and roughly on par with Sonnet 5.5.

2026-10-24 ~ 2026-10-24 · 2 related posts