/ Models Index / Kimi K2

Kimi K2

Frontier · Moonshot · Moonshot AI · Released Q2 2025
B+

Good (B+) — Capability Grade

Kimi K2 is Moonshot AI's flagship — Chinese-headquartered, competitive with Western frontier models on long-context and Chinese-language benchmarks. 2M context window matches Gemini.

84
Composite / 100
/ Subscore Breakdown · 6 Capability Dimensions

Where this grade comes from.

General Reasoning
B+
Code Generation
B
Math & STEM
B+
Tool Use & Agency
B
Multimodal
B
Safety & Alignment
B
/ Key Events & Disclosures

Release timeline & positioning.

  • Released Q2 2025
  • Chinese-headquartered (Moonshot)
  • 2M context window
  • Competitive long-context

/ Best for

Chinese-language workloads and long-context applications.

/ Watch out for

China-headquartered may produce regulatory considerations for some enterprises and government deployments.