Introducing Kimi K3.1 Pro
Kimi K3.1 Pro is the first near-frontier open-weight model, scoring 55 on Artificial Analysis
“This is the first time a (soon-to-be) open weight model has scored this highly against the frontier.”
Moonshot AI launched Kimi K3.1 Pro on its API today, with open weights promised by November 4, 2026. The model scores 55 on Artificial Analysis — ahead of Claude Fable 5.1, GPT-6 Astra, and Gemini 4 Argon — making it the highest-scoring open-weight model to date. At $2/M input and $10/M output tokens, it undercuts Claude Opus 5.5 by half, raising the competitive stakes for both capability and cost among frontier models.