Moonshot AILanguage modelSupersededOpen weights
Kimi K2 Thinking
Open reasoning agent that could chain hundreds of tool calls.
Overall Intelligence
#38
of 56
18.5 pts behind #1
0
races led
4
weeks at #1
70.3
score
- Released
- Nov 6, 2025
- 12 mo ago
- Context
- 256K
- 64K max output
- Price
- $0.6 / $2.5
- per 1M in/out · $1.07 blended
- Speed
- 50 tok/s
- 6s to first token
- Modalities
- text
- outputs text
- Weights
- Open
- 1T MoE (32B active)
The 30-second version
Open reasoning agent that could chain hundreds of tool calls.
Key takeaways
- 1Open reasoning agent that could chain hundreds of tool calls.
Where it ranks
Current snapshot · illustrative scores · click any race to watch it
K2 ThinkingBest on each axis
Overall
Coding
37Coding+18.3 pts41Frontend & UI+123 Elo37Backend & APIs+16.5 pts37Distributed Systems+18.4 pts35Full-Stack Apps+17.7 pts36Debugging & Repair+18.1 pts34Agentic Coding+18.1 pts36Code Review & Security+15.6 pts44Data Science & SQL+20.0 pts40Mobile Apps+17.2 pts30Terminal & DevOps+18.6 pts40Competitive Programming+529
Hard Reasoning
Math
Science
Writing & Language
Multimodal
Tool Use & Agents
Speed & Cost
Overall Intelligence over time
#1 frontierK2 ThinkingV4.1 FlashGLM-5.3
Update history
The distilled version first; open any release for the deeper breakdown.
The Kimi family