Neura Intelligence Index
Category Comparison
Benchmark-by-Benchmark
coding
| Benchmark | MiMo-V2.5 | Mistral Large 2 |
|---|---|---|
| SWE-Bench Pro | 56.1 | — |
| HumanEval | — | 92 |
multimodal
| Benchmark | MiMo-V2.5 | Mistral Large 2 |
|---|---|---|
| MMMU-Pro | 77.9 | — |
| CharXiv Reasoning | 81 | — |
language
| Benchmark | MiMo-V2.5 | Mistral Large 2 |
|---|---|---|
| HellaSwag | — | 93.6 |
safety
| Benchmark | MiMo-V2.5 | Mistral Large 2 |
|---|---|---|
| TruthfulQA | — | 60.5 |
reasoning
| Benchmark | MiMo-V2.5 | Mistral Large 2 |
|---|---|---|
| GPQA Diamond | — | 48.2 |
| ARC-Challenge | — | 94 |
general
| Benchmark | MiMo-V2.5 | Mistral Large 2 |
|---|---|---|
| MMLU-Pro | — | 69.4 |
| IFEval | — | 82.8 |
| Chatbot Arena Elo | — | 1250 |
Category Winners
coding
80.3
math
46.2
reasoning
35.9
general
55.9
language
66.8
multimodal
X
MiMo-V2.571.3
safety
29.8