Neura Intelligence Index
AI Model Benchmarks
| Benchmark | Kimi K2.5 | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 87.6 | 90.5 |
| Humanity's Last Exam | 50.2 | 36.4 |
| Benchmark | Kimi K2.5 | Kimi K2.6 |
|---|---|---|
| AIME 2025 | 96.1 | — |
| Benchmark | Kimi K2.5 | Kimi K2.6 |
|---|---|---|
| SWE-bench Verified | 76.8 | 80.2 |
| SWE-Bench Pro | 50.7 | 58.6 |
| SciCode | 48.7 | 52.2 |
| Benchmark | Kimi K2.5 | Kimi K2.6 |
|---|---|---|
| MMMU-Pro | 78.5 | 80.1 |
| CharXiv Reasoning | 77.5 | 86.7 |
| Benchmark | Kimi K2.5 | Kimi K2.6 |
|---|---|---|
| BrowseComp | 74.9 | 86.3 |
| Toolathlon | — | 50 |
| APEX Agents | — | 27.9 |