# MiMo-V2.5 vs Qwen3 14B

> MiMo-V2.5 is the stronger model overall, scoring 43.0 to 35.0 on the Noometry Index.

- Canonical page: https://noometry.com/compare/mimo-v2-5-vs-qwen3-14b
- Last updated: 2026-10-10
- Shared benchmarks: 2

## Summary

- They share 2 benchmarks with published results for both. MiMo-V2.5 scores higher in 4 categories and Qwen3 14B in 1 category; 4 gaps are clear of the uncertainty.
- The widest gap is in reasoning, where MiMo-V2.5 leads 27.8 to 18.1.
- The biggest single-benchmark swing is SciCode: 43.1% for MiMo-V2.5 and 31.6% for Qwen3 14B.
- MiMo-V2.5 is cheaper at $0.14 / $0.28 per million input/output tokens, against $0.35 / $1.40 for Qwen3 14B.
- MiMo-V2.5 accepts more context: 1.05M tokens versus 131K.

## Snapshot

| | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| Provider | Xiaomi | Alibaba (Qwen) |
| Noometry Index | 43.0 | 35.0 |
| Rank | 89 | 222 |
| Context | 1.05M | 131K |
| Input $/M | $0.14 | $0.35 |
| Output $/M | $0.28 | $1.40 |
| Weights | Open | Open |

## Coding

- MiMo-V2.5: 43.6 (#75)
- Qwen3 14B: 36.7 (#191)

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| SciCode | 43.1% | 31.6% |
| LMArena WebDev | 1438 | — |
| LMArena Coding | 1469 | — |
| ALE-Bench | 513.95 | — |

## Agentic & Tool Use

- MiMo-V2.5: —
- Qwen3 14B: 29.4 (#78)

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| Berkeley Function Calling Leaderboard | — | 41% |

## Reasoning

- MiMo-V2.5: 27.8 (#98)
- Qwen3 14B: 18.1 (#272)

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| CritPt | 3.7% | 0% |
| Kagi LLM Benchmark | — | 49.1% |
| Chess Puzzles | — | 4% |
| LMArena Hard Prompts | 1450 | — |
| DTBench | — | 64% |
| LMCA | — | 18.2% |
| Epoch Capabilities Index | — | 138.23 |

## Math

- MiMo-V2.5: 35.4 (#145)
- Qwen3 14B: 37.0 (#102)

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| OTIS Mock AIME 2024-2025 | — | 66.4% |
| ProofBench | 16% | — |
| LMArena Math | 1436 | — |

## Knowledge

- MiMo-V2.5: 40.2 (#109)
- Qwen3 14B: 39.2 (#124)

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| GPQA Diamond | — | 63.8% |
| Vectara Hallucination Rate | — | 5.4% |
| LMArena Expert | 1459 | — |

## Multimodal

- MiMo-V2.5: 39.0 (#53)
- Qwen3 14B: —

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| LMArena Vision | 1245 | — |

## Multilingual

- MiMo-V2.5: 52.0 (#96)
- Qwen3 14B: —

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| LMArena Non-English | 1404 | — |
| LMArena Chinese | 1467 | — |
| LMArena French | 1447 | — |
| LMArena German | 1421 | — |
| LMArena Japanese | 1304 | — |
| LMArena Korean | 1363 | — |
| LMArena Russian | 1395 | — |
| LMArena Spanish | 1412 | — |

## Instruction Following

- MiMo-V2.5: 75.7 (#55)
- Qwen3 14B: —

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| LMArena Instruction Following | 1433 | — |

## Long Context

- MiMo-V2.5: 44.1 (#69)
- Qwen3 14B: 38.0 (#201)

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| Fiction.LiveBench | — | 62.5% |
| LMArena Longer Query | 1444 | — |

## Writing & Preference

- MiMo-V2.5: 61.8 (#82)
- Qwen3 14B: —

| Benchmark | MiMo-V2.5 | Qwen3 14B |
|---|---|---|
| LMArena Text | 1427 | — |
| LMArena Creative Writing | 1392 | — |
| LMArena Multi-Turn | 1445 | — |

## FAQ

### Is MiMo-V2.5 better than Qwen3 14B?

MiMo-V2.5 is the stronger model overall, scoring 43.0 to 35.0 on the Noometry Index.

### Which is cheaper, MiMo-V2.5 or Qwen3 14B?

MiMo-V2.5 is cheaper. It lists at $0.14 per million input tokens and $0.28 per million output tokens; Qwen3 14B lists at $0.35 and $1.40.

### Is MiMo-V2.5 or Qwen3 14B better for coding?

MiMo-V2.5 scores higher on coding benchmarks: 43.6 versus 36.7 in the Noometry coding category.

### Which has the bigger context window?

MiMo-V2.5 does, with 1.05M tokens against 131K.

### How many benchmarks do MiMo-V2.5 and Qwen3 14B share?

2 benchmarks have published results for both models. MiMo-V2.5 has 23 scored results on Noometry and Qwen3 14B has 12.