LLMOpen weights
Kimi K2.6
Moonshot AI
Updated May 21, 2026 · Benchmarks via Artificial Analysis, specs & pricing via OpenRouter · Methodology · Report an error
Specifications
- Type
- LLM
- Access
- Open weights
- Released
- April 20, 2026
- License
- Modified MIT
- Parameters
- 1T (32B active)
- Context window
- 262,144 tokens
- Output speed
- 57 tok/s
- Latency (TTFT)
- 1.2s
- Input
- Text, Image
- Output
- Text
- API pricing
- $0.73 in · $3.49 out / 1M tokens
Capabilities
Function callingStructured output
Benchmarks
Reasoning
| GPQA Diamond | 91.1 |
Coding
| SWE-bench Pro | 58.6 |
General
| Humanity's Last Exam (w/ tools) | 54 |
| Humanity’s Last Exam | 35.9 |
Other
| Humanity's Last Exam (w/ tools) | 54 |
Our take
Latest Kimi flagship; ties top closed coding models on SWE-bench Pro and leads several agentic benchmarks among open models.