MiMo-V2.5-Pro (Non-reasoning) — benchmark results
MiMo-V2.5-Pro evaluated with reasoning disabled. Provider: Xiaomi. Released 2026-04-23. Access: Open.
Unified ELO 1690 ± 15, rank #292 of 1776 rated models, from 35 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| UGI - Writing | 51.71 | Writing Score | 89.7 |
| UGI - Natural Intelligence | 41.05 | NatInt Score | 87.7 |
| AA Omniscience - Software Engineering (SWE) - Kotlin | 44 | Accuracy (%) | 85.7 |
| AA Omniscience - Software Engineering (SWE) - R | 40 | Accuracy (%) | 85.5 |
| AA Omniscience - Software Engineering (SWE) - HTML | 54 | Accuracy (%) | 83.3 |
| AA Terminal-Bench Hard | 35.61 | Accuracy (%) | 82 |
| AA Omniscience - Software Engineering (SWE) - Go | 36 | Accuracy (%) | 80.6 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 45.56 | Accuracy (%) | 78.7 |
| AA Omniscience - Software Engineering (SWE) - Swift | 56 | Accuracy (%) | 77.6 |
| AA Omniscience - Software Engineering (SWE) - Java | 28 | Accuracy (%) | 77.5 |
| AA Omniscience - Software Engineering (SWE) | 39.5 | Accuracy (%) | 77 |
| AA Omniscience - Software Engineering (SWE) - PHP | 40 | Accuracy (%) | 75 |
Interactive version: theaggregate.ai/model?slug=mimo-v2-5-pro-non-reasoning · How the rankings work · Data refreshed daily, snapshot 2026-07-22.