Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | GPT-5.5 Pro | MiMo-V2-Pro |
|---|---|---|
| Humanity's Last Exam | 57.2 | — |
agent
| Benchmark | GPT-5.5 Pro | MiMo-V2-Pro |
|---|---|---|
| BrowseComp | 90.1 | — |
math
| Benchmark | GPT-5.5 Pro | MiMo-V2-Pro |
|---|---|---|
| FrontierMath | 39.6 | — |
coding
| Benchmark | GPT-5.5 Pro | MiMo-V2-Pro |
|---|---|---|
| SWE-bench Verified | — | 78 |
Category Winners
coding
X
MiMo-V2-Pro77.6
math
57.5
reasoning
91.8
agent
89.2