Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Mistral Small 3.1 24B Base | Qwen2.5 VL 32B Instruct |
|---|---|---|
| GPQA Diamond | 37.5 | 46 |
multimodal
| Benchmark | Mistral Small 3.1 24B Base | Qwen2.5 VL 32B Instruct |
|---|---|---|
| MMMU | 59.3 | 70 |
| MMMU-Pro | — | 49.5 |
| ScreenSpot Pro | — | 39.4 |
agent
| Benchmark | Mistral Small 3.1 24B Base | Qwen2.5 VL 32B Instruct |
|---|---|---|
| OSWorld | — | 5.9 |
Category Winners
reasoning
Q
Qwen2.5 VL 32B Instruct13.3
multimodal
26.7
agent
Q
Qwen2.5 VL 32B Instruct2.8