MiMo-V2-Flash (Non-reasoning): benchmark results
MiMo-V2-Flash non-reasoning evaluation mode. Provider: Xiaomi. Released 2025-12-16. Access: Open.
Aggregate Rating 1516 ± 14, rank #1010 of 2168 rated models, from 135 benchmark results.
Strongest benchmark results
| Benchmark | Score | Metric | Percentile |
|---|---|---|---|
| AA Omniscience - Software Engineering (SWE) - Dart | 28 | Accuracy (%) | 82.1 |
| UGI - Writing | 41.64 | Writing Score | 76.9 |
| AA Omniscience - Software Engineering (SWE) - R | 18 | Accuracy (%) | 76.8 |
| Tau2-Bench Telecom | 83.92 | Success Rate (%) | 75.1 |
| AA TAU-2 Bench | 83.92 | Accuracy (%) | 74.8 |
| UGI - Natural Intelligence | 30.72 | NatInt Score | 74.2 |
| AA Omniscience - Software Engineering (SWE) - Julia | 16 | Accuracy (%) | 68.8 |
| AA Terminal-Bench Hard | 25.76 | Accuracy (%) | 67.1 |
| AA Omniscience - Software Engineering (SWE) - TypeScript | 25.56 | Accuracy (%) | 65.9 |
| AA Omniscience - Software Engineering (SWE) - Python | 22.5 | Accuracy (%) | 65.7 |
| AA Omniscience - Software Engineering (SWE) - JavaScript | 31.82 | Accuracy (%) | 64.3 |
| Chatbot Arena (Text - Hard Prompts (English)) | 1434 | Arena Score | 62.9 |
Interactive version: theaggregate.ai/model?slug=mimo-v2-flash-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-10-20.