Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
math
| Benchmark | Mistral Medium 3.5 | Qwen3.6 Plus |
|---|---|---|
| AIME 2025 | 86.3 | — |
coding
| Benchmark | Mistral Medium 3.5 | Qwen3.6 Plus |
|---|---|---|
| SWE-bench Verified | 77.6 | 78.8 |
| SWE-Bench Pro | — | 56.6 |
agent
| Benchmark | Mistral Medium 3.5 | Qwen3.6 Plus |
|---|---|---|
| BrowseComp | 48.6 | — |
| Toolathlon | — | 39.8 |
| MCP Atlas | — | 74.1 |
reasoning
| Benchmark | Mistral Medium 3.5 | Qwen3.6 Plus |
|---|---|---|
| GPQA Diamond | — | 90.4 |
| Humanity's Last Exam | — | 28.8 |
multimodal
| Benchmark | Mistral Medium 3.5 | Qwen3.6 Plus |
|---|---|---|
| MMMU | — | 86 |
| MMMU-Pro | — | 78.8 |
| CharXiv Reasoning | — | 81.5 |
| ScreenSpot Pro | — | 68.2 |
general
| Benchmark | Mistral Medium 3.5 | Qwen3.6 Plus |
|---|---|---|
| MMMLU | — | 89.5 |
Category Winners
coding
77.0
math
63.4
reasoning
Q
Qwen3.6 Plus66.7
general
Q
Qwen3.6 Plus73.8
multimodal
Q
Qwen3.6 Plus75.8
agent
Q
Qwen3.6 Plus41.0