MiMo-V2.5-Pro (Non-reasoning): benchmark results
MiMo-V2.5-Pro evaluated with reasoning disabled. Provider: Xiaomi. Released 2026-04-23. Access: Open.
Aggregate Rating 1660 ± 18, rank #490 of 2168 rated models, from 61 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| RideWay - Success Rate | 87.4 | Success rate (%; share of trials passing the three-judge LLM | 100 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 46 | Accuracy (%) | 96.1 |
| AA Omniscience - Software Engineering (SWE) - R | 40 | Accuracy (%) | 95.6 |
| AA Omniscience - Software Engineering (SWE) - HTML | 56 | Accuracy (%) | 93.9 |
| AA Omniscience - Software Engineering (SWE) - Go | 38 | Accuracy (%) | 91.5 |
| VEHBench - Corrupted-State Recovery | 45.5 | P3 success (%; share of corrupted or misleading design traje | 90.9 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 45.56 | Accuracy (%) | 89.6 |
| AA Omniscience - Software Engineering (SWE) - Java | 28 | Accuracy (%) | 88.1 |
| AA Omniscience - Software Engineering (SWE) - Dart | 32 | Accuracy (%) | 87.4 |
| UGI - Writing | 51.71 | Writing Score | 87.3 |
| AA Omniscience - Software Engineering (SWE) - Swift | 52 | Accuracy (%) | 86.7 |
| UGI - Natural Intelligence | 41.05 | NatInt Score | 85.5 |
Interactive version: theaggregate.ai/model?slug=mimo-v2-5-pro-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.