Xiaomi
Xiaomi MiMo-V2-Flash is a proprietary MoE model developed by Xiaomi, designed for extreme inference efficiency with 309B total parameters (15B active). By incorporating an innovative Hybrid attention architecture and multi-layer MTP inference acceleration, it ranks among the top 2 global open-source models across multiple Agent benchmarks.
Xiaomi model releasesRank #80 across 526
Rank #118 across 436
Rank #48 across 357
Percentile score by analysis domain.
* Cost is inverted: lower input, output, and blended prices rank higher.
Higher bars mean stronger relative placement.
| Metric | Domain | Value | Rank |
|---|---|---|---|
| Artificial Analysis Intelligence Index | overall | 39.2 | #80 |
| Artificial Analysis Coding Index | coding | 31.8 | #118 |
| Artificial Analysis Math Index | math | 96.3 | #6 |
| MMLU-Pro | reasoning | 84.3% | #38 |
| reasoning |
| 84.6% |
| #68 |
| Humanity's Last Exam | reasoning | 21.1% | #74 |
| LiveCodeBench | coding | 86.8% | #10 |
| SciCode | coding, reasoning | 39.4% | #127 |
| Output Speed | speed | 129.5 tok/s | #101 |
| Time to First Token | speed | 1.29s | #171 |
| Blended Price | cost | $0.150/M | #48 |
| Input Price | cost | $0.100/M | #52 |
| Output Price | cost | $0.300/M | #55 |
| Value Index | cost, overall | 261.3 | #11 |