Introduction

The large language model (LLM) race is heating up fast — and three models are leading the conversation: MiniMax M2, GLM 4.6, and Kimi-K2-Thinking.

Introduction

Recently, I’ve been exploring large language models (LLMs) and came across Moonshot AI, a company that really caught my attention. Their flagship model Kimi K2 claims to handle up to 128K tokens of context and uses a cutting-edge Mixture of Experts (MoE) architecture.