Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Claude Opus 4.5 | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 87 | 90.5 |
| ARC-AGI v2 | 37.6 | — |
| Humanity's Last Exam | — | 36.4 |
coding
| Benchmark | Claude Opus 4.5 | Kimi K2.6 |
|---|---|---|
| SWE-bench Verified | 80.9 | 80.2 |
| SWE-Bench Pro | — | 58.6 |
| SciCode | — | 52.2 |
general
| Benchmark | Claude Opus 4.5 | Kimi K2.6 |
|---|---|---|
| MMMLU | 90.8 | — |
agent
| Benchmark | Claude Opus 4.5 | Kimi K2.6 |
|---|---|---|
| OSWorld | 66.3 | — |
| MCP Atlas | 62.3 | — |
| BrowseComp | — | 86.3 |
| Toolathlon | — | 50 |
| APEX Agents |
multimodal
| Benchmark | Claude Opus 4.5 | Kimi K2.6 |
|---|---|---|
| MMMU-Pro | — | 80.1 |
| CharXiv Reasoning | — | 86.7 |
Category Winners
coding
82.6
reasoning
M
Kimi K2.673.3
general
78.5
multimodal
M
Kimi K2.680.4
agent
52.1