← Models
Kimi K2.5
Moonshot AI · open weight · kimi-k2.5
Kimi K2.5 is Moonshot AI's flagship agentic model and a new SOTA open model. It unifies vision and text, thinking and non-thinking modes, and single-agent and multi-agent execution into one model. Built with Full-Parameter RL tuning, it achieves state-of-the-art performance across agents, coding, image, and video benchmarks.
Benchmark scores
| Benchmark | Score |
|---|---|
| AIME 2025 | 0.96 |
| HMMT 2025 | 0.95 |
| InfoVQAtest | 0.93 |
| OCRBench | 0.92 |
| MathVista-Mini | 0.90 |
| OmniDocBench 1.5 | 0.89 |
| GPQA | 0.88 |
| Video-MME | 0.87 |
| MMLU-Pro | 0.87 |
| VideoMMMU | 0.87 |
| WideSearch | 0.79 |
| MMMU-Pro | 0.79 |
| DeepSearchQA | 0.77 |
| SWE-Bench Verified | 0.77 |
| BrowseComp | 0.75 |
| SWE-bench Multilingual | 0.73 |
| Humanity's Last Exam | 0.50 |
Pricing
- No provider pricing.
AA metrics
No Artificial Analysis link yet.
Arena Elo
- No Arena snapshot linked.