Kimi K2 Thinking
MoonshotAiBalanced
ThinkingTool UseStructured Output
About this model
Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in...
Performance Tier
Balanced
Kimi K2 Thinking is a balanced model from MoonshotAi : strong performance at a reasonable price.
Strong cost-performance ratio. Reliable for most professional use cases without premium pricing.
Pricing
This model is included in Elosia plans| Type | per 1M tokens |
|---|---|
| Input (prompt) | $0.600 |
| Output (completion) | $2.50 |
Capabilities
Context Length262K
Max Output Tokens—
TokenizerOther
Inputtext
Outputtext
Release DateNovember 6, 2025
Benchmarks
General Intelligence
MMLU
84%
GPQA Diamond
60.5%
Mathematics
MATH-500
88%
Programming
HumanEval
84%
Recommended Use Cases
MathematicsAnalysisResearchCoding
Strengths
- Extended thinking mode for complex reasoning tasks
- Strong mathematical reasoning capabilities
- Transparent chain-of-thought process
Limitations
- Slower than standard Kimi K2 due to thinking overhead
- Less suited for quick conversational interactions
Resources
This model may use your data for training
