Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | Claude Opus 4 | Kimi K2.6 |
|---|---|---|
| HumanEval | 95.1 | — |
| SWE-bench Verified | 55.8 | 80.2 |
| SWE-Bench Pro | — | 58.6 |
| SciCode | — | 52.2 |
reasoning
| Benchmark | Claude Opus 4 | Kimi K2.6 |
|---|---|---|
| GPQA Diamond | 74.8 | 90.5 |
| BigBench-Hard | 93.2 | — |
| Humanity's Last Exam | — | 36.4 |
general
| Benchmark | Claude Opus 4 | Kimi K2.6 |
|---|---|---|
| IFEval | 91.5 | — |
| MMLU-Pro | 83.5 | — |
| Chatbot Arena Elo | 1365 | — |
multimodal
| Benchmark | Claude Opus 4 | Kimi K2.6 |
|---|---|---|
| MMMU | 74.2 | — |
| MMMU-Pro | — | 80.1 |
| CharXiv Reasoning | — | 86.7 |
agent
| Benchmark | Claude Opus 4 | Kimi K2.6 |
|---|---|---|
| BrowseComp | — | 86.3 |
| Toolathlon | — | 50 |
| APEX Agents | — | 27.9 |
Category Winners
coding
M
Kimi K2.671.2
math
75.5
reasoning
80.3
general
96.0
multimodal
M
Kimi K2.682.2
agent
M
Kimi K2.655.0