Kimi K2 Thinking
MoonshotLanguage modeling/generationQuestion answeringSearchQuantitative reasoningCode generationOpen weights
Developed by Moonshot in 2025, Kimi K2 Thinking is a language modeling/generation model with 1000000000000.0 parameters with openly available weights.
About Kimi K2 Thinking
Today, we are introducing Kimi K2 Thinking, our best open-source thinking model.
Built as a thinking agent, it reasons step by step while using tools, achieving state-of-the-art performance on Humanity's Last Exam (HLE), BrowseComp, and other benchma
LLM pricing & performance
Full LLM page →Kimi K2 Thinking is available via API — live cost, context, and benchmark data:
Input / 1M
$0.00
Output / 1M
$0.00
Context
262K
Tokens/sec
—
Details
- Provider
- Moonshot
- Task
- Language modeling/generation,Question answering,Search,Quantitative reasoning,Code generation
- Parameters
- 1000000000000.0
- Released
- 2025-11-06
- Open weights
- Yes