Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemma 4 12B | QwQ-32B-Preview |
|---|---|---|
| GPQA Diamond | 78.8 | 65.2 |
| Humanity's Last Exam | 5.2 | — |
multimodal
| Benchmark | Gemma 4 12B | QwQ-32B-Preview |
|---|---|---|
| MMMU-Pro | 69.1 | — |
Category Winners
reasoning
Q
QwQ-32B-Preview47.0
general
50.0
multimodal
53.2