Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
reasoning
| Benchmark | Gemini 3.1 Flash-Lite | MiMo-V2-Flash |
|---|---|---|
| GPQA Diamond | 86.9 | 83.7 |
| Humanity's Last Exam | 16 | 22.1 |
multimodal
| Benchmark | Gemini 3.1 Flash-Lite | MiMo-V2-Flash |
|---|---|---|
| MMMU-Pro | 76.8 | — |
| CharXiv Reasoning | 73.2 | — |
math
| Benchmark | Gemini 3.1 Flash-Lite | MiMo-V2-Flash |
|---|---|---|
| AIME 2025 | — | 94.1 |
coding
| Benchmark | Gemini 3.1 Flash-Lite | MiMo-V2-Flash |
|---|---|---|
| SWE-bench Verified | — | 73.4 |
| TerminalBench | — | 30.5 |
agent
| Benchmark | Gemini 3.1 Flash-Lite | MiMo-V2-Flash |
|---|---|---|
| BrowseComp | — | 58.3 |
Category Winners
coding
X
MiMo-V2-Flash51.1
math
X
MiMo-V2-Flash76.5
reasoning
X
MiMo-V2-Flash53.8
general
65.0
multimodal
58.9
agent
X
MiMo-V2-Flash36.0