Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Kimi K2-Thinking-0905 | Midjourney v6.1 |
|---|---|---|
| GPQA Diamond | 84.5 | — |
| Humanity's Last Exam | 51 | — |
math
| Benchmark | Kimi K2-Thinking-0905 | Midjourney v6.1 |
|---|---|---|
| AIME 2025 | 100 | — |
coding
| Benchmark | Kimi K2-Thinking-0905 | Midjourney v6.1 |
|---|---|---|
| SWE-bench Verified | 71.3 | — |
| TerminalBench | 47.1 | — |
| SciCode | 44.8 | — |
agent
| Benchmark | Kimi K2-Thinking-0905 | Midjourney v6.1 |
|---|---|---|
| BrowseComp | 60.2 | — |
image_quality
| Benchmark | Kimi K2-Thinking-0905 | Midjourney v6.1 |
|---|---|---|
| Image Arena Elo | — | 1215 |
Category Winners
coding
M
Kimi K2-Thinking-090566.5
math
M
Kimi K2-Thinking-090584.3
reasoning
M
Kimi K2-Thinking-090582.6
agent
M
Kimi K2-Thinking-090539.6