Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | MAI-Thinking-1 | Qwen3.5-122B-A10B |
|---|---|---|
| GPQA Diamond | 84.2 | 86.6 |
| Humanity's Last Exam | — | 47.5 |
math
| Benchmark | MAI-Thinking-1 | Qwen3.5-122B-A10B |
|---|---|---|
| AIME 2025 | 97 | — |
coding
| Benchmark | MAI-Thinking-1 | Qwen3.5-122B-A10B |
|---|---|---|
| SWE-bench Verified | 73.5 | 72 |
| SWE-Bench Pro | 52.8 | — |
multimodal
| Benchmark | MAI-Thinking-1 | Qwen3.5-122B-A10B |
|---|---|---|
| MMMU | — | 83.9 |
| MMMU-Pro | — | 76.9 |
| CharXiv Reasoning | — | 77.2 |
| ScreenSpot Pro | — | 70.4 |
general
| Benchmark | MAI-Thinking-1 | Qwen3.5-122B-A10B |
|---|---|---|
| MMMLU | — | 86.7 |
agent
| Benchmark | MAI-Thinking-1 | Qwen3.5-122B-A10B |
|---|---|---|
| BrowseComp | — | 63.8 |
Category Winners
coding
Q
Qwen3.5-122B-A10B65.2
math
80.6
reasoning
Q
Qwen3.5-122B-A10B81.5
general
Q
Qwen3.5-122B-A10B64.0
multimodal
Q
Qwen3.5-122B-A10B70.3
agent
Q
Qwen3.5-122B-A10B47.2