Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
math
| Benchmark | Qwen3 VL 235B A22B Thinking | Wan2.1 |
|---|---|---|
| AIME 2025 | 89.7 | — |
reasoning
| Benchmark | Qwen3 VL 235B A22B Thinking | Wan2.1 |
|---|---|---|
| Humanity's Last Exam | 13.6 | — |
multimodal
| Benchmark | Qwen3 VL 235B A22B Thinking | Wan2.1 |
|---|---|---|
| MMMU-Pro | 69.3 | — |
| CharXiv Reasoning | 66.1 | — |
| ScreenSpot Pro | 61.8 | — |
general
| Benchmark | Qwen3 VL 235B A22B Thinking | Wan2.1 |
|---|---|---|
| SimpleQA | 44.4 | — |
agent
| Benchmark | Qwen3 VL 235B A22B Thinking | Wan2.1 |
|---|---|---|
| OSWorld | 38.1 | — |
video_quality
| Benchmark | Qwen3 VL 235B A22B Thinking | Wan2.1 |
|---|---|---|
| VBench | — | 79.3 |
Category Winners
math
Q
Qwen3 VL 235B A22B Thinking69.5
reasoning
Q
Qwen3 VL 235B A22B Thinking15.4
general
Q
Qwen3 VL 235B A22B Thinking59.9
multimodal
Q
Qwen3 VL 235B A22B Thinking41.1
agent
Q
Qwen3 VL 235B A22B Thinking35.7