Neura Intelligence Index
| Benchmark | Phi 4 Reasoning | Qwen3 VL 4B Thinking |
|---|---|---|
| GPQA Diamond | 65.8 | 64.1 |
| Benchmark | Phi 4 Reasoning | Qwen3 VL 4B Thinking |
|---|---|---|
| AIME 2025 | 62.9 | 74.5 |
| Benchmark | Phi 4 Reasoning | Qwen3 VL 4B Thinking |
|---|---|---|
| MMMU-Pro | — | 57 |
| CharXiv Reasoning | — | 50.3 |
| ScreenSpot Pro | — | 49.2 |
| Benchmark | Phi 4 Reasoning | Qwen3 VL 4B Thinking |
|---|---|---|
| OSWorld | — | 31.4 |