Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | o1 | Qwen3 VL 235B A22B Thinking |
|---|---|---|
| GPQA Diamond | 78 | — |
| Humanity's Last Exam | — | 13.6 |
coding
| Benchmark | o1 | Qwen3 VL 235B A22B Thinking |
|---|---|---|
| SWE-bench Verified | 41 | — |
multimodal
| Benchmark | o1 | Qwen3 VL 235B A22B Thinking |
|---|---|---|
| MMMU | 77.6 | — |
| MMMU-Pro | — | 69.3 |
| CharXiv Reasoning | — | 66.1 |
| ScreenSpot Pro | — | 61.8 |
agent
| Benchmark | o1 | Qwen3 VL 235B A22B Thinking |
|---|---|---|
| TAU-Bench Retail | 70.8 | — |
| OSWorld | — | 38.1 |
math
| Benchmark | o1 | Qwen3 VL 235B A22B Thinking |
|---|---|---|
| FrontierMath | 5.5 | — |
| AIME 2025 | — | 89.7 |
Category Winners
coding
6.8
math
Q
Qwen3 VL 235B A22B Thinking69.5
reasoning
69.8
general
65.6
multimodal
75.8
agent
59.1