Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemma 4 31B | Qwen3.5-397B-A17B |
|---|---|---|
| GPQA Diamond | 84.3 | 88.4 |
| Humanity's Last Exam | 26.5 | 28.7 |
multimodal
| Benchmark | Gemma 4 31B | Qwen3.5-397B-A17B |
|---|---|---|
| MMMU-Pro | 76.9 | — |
coding
| Benchmark | Gemma 4 31B | Qwen3.5-397B-A17B |
|---|---|---|
| SWE-bench Verified | — | 76.4 |
agent
| Benchmark | Gemma 4 31B | Qwen3.5-397B-A17B |
|---|---|---|
| BrowseComp | — | 69 |
| Toolathlon | — | 38.3 |
Category Winners
coding
Q
Qwen3.5-397B-A17B74.2
reasoning
Q
Qwen3.5-397B-A17B63.8
general
71.5
multimodal
73.4
agent
Q
Qwen3.5-397B-A17B36.5