Moonshot’s Kimi K3 tops frontend coding but lags in expert math

Moonshot AI's Kimi K3 has claimed the top position on the Code Arena Frontend leaderboard with 1,679 points based on human preference ratings. It surpasses Claude Fable 5 at 1,631 points and GPT-5.6 Sol at 1,618 points, becoming the first Chinese model to lead this ranking. The model is a 2.8-trillion-parameter sparse mixture-of-experts system with a 1-million-token context window. Source

Les hele artikkelen hos kilden.