Kimi K3 reveals its pre-training mix across text, code, math, knowledge, and vision

stochasticchasm · x · 2026-07-28

Kimi K3’s pre-training data is described as a curated mix of four main text domains — web text, code, mathematics, and knowledge — plus a large-scale vision corpus.

Related event: Inside Kimi K3: Tri-axis Architecture and Hybrid Attention(34 posts)→

Original post →

More from Models

Models channel →