Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemma 4 31B | Qwen3.5-122B-A10B |
|---|---|---|
| GPQA Diamond | 84.3 | 86.6 |
| Humanity's Last Exam | 26.5 | 47.5 |
multimodal
| Benchmark | Gemma 4 31B | Qwen3.5-122B-A10B |
|---|---|---|
| MMMU-Pro | 76.9 | 76.9 |
| MMMU | — | 83.9 |
| CharXiv Reasoning | — | 77.2 |
| ScreenSpot Pro | — | 70.4 |
coding
| Benchmark | Gemma 4 31B | Qwen3.5-122B-A10B |
|---|---|---|
| SWE-bench Verified | — | 72 |
agent
| Benchmark | Gemma 4 31B | Qwen3.5-122B-A10B |
|---|---|---|
| BrowseComp | — | 63.8 |
Category Winners
coding
Q
Qwen3.5-122B-A10B65.6
reasoning
Q
Qwen3.5-122B-A10B83.0
general
76.8
multimodal
73.9
agent
Q
Qwen3.5-122B-A10B47.5