Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | o3 | Qwen3 VL 8B Thinking |
|---|---|---|
| GPQA Diamond | 83.3 | 69.9 |
| Humanity's Last Exam | 14.7 | — |
| ARC-AGI v2 | 6.5 | — |
math
| Benchmark | o3 | Qwen3 VL 8B Thinking |
|---|---|---|
| AIME 2025 | 86.4 | 80.3 |
| FrontierMath | 15.8 | — |
coding
| Benchmark | o3 | Qwen3 VL 8B Thinking |
|---|---|---|
| SWE-bench Verified | 69.1 | — |
multimodal
| Benchmark | o3 | Qwen3 VL 8B Thinking |
|---|---|---|
| MMMU | 82.9 | — |
| MMMU-Pro | 76.4 | 60.4 |
| CharXiv Reasoning | 78.6 | 53 |
| ScreenSpot Pro | — | 46.6 |
agent
| Benchmark | o3 | Qwen3 VL 8B Thinking |
|---|---|---|
| BrowseComp | 49.7 | — |
| OSWorld | — | 33.9 |
general
| Benchmark | o3 | Qwen3 VL 8B Thinking |
|---|---|---|
| SimpleQA | — | 49.6 |
Category Winners
coding
57.9
math
Q
Qwen3 VL 8B Thinking52.0
reasoning
Q
Qwen3 VL 8B Thinking53.6
general
Q
Qwen3 VL 8B Thinking67.5
multimodal
72.5
agent
Q
Qwen3 VL 8B Thinking27.4