# Mixtral 8x22B vs Qwen2.5-Max

> Qwen2.5-Max is the stronger model overall, scoring 40.1 to 26.7 on the Noometry Index.

- Canonical page: https://noometry.com/compare/mixtral-8x22b-vs-qwen2-5-max
- Last updated: 2026-10-10
- Shared benchmarks: 18

## Summary

- They share 18 benchmarks with published results for both. Mixtral 8x22B scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen2.5-Max leads 34.8 to 15.1.
- Mixtral 8x22B has downloadable open weights; the other is API-only.

## Snapshot

| | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 26.7 | 40.1 |
| Rank | 330 | 142 |
| Context | 64K | — |
| Input $/M | $2 | — |
| Output $/M | $6 | — |
| Weights | Open | Proprietary |

## Coding

- Mixtral 8x22B: 23.8 (#324)
- Qwen2.5-Max: 41.2 (#118)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1166 | 1359 |
| WeirdML | 3.2% | — |
| BigCodeBench Instruct | 40.6% | — |
| LiveBench Coding | — | 64.4% |
| BigCodeBench Complete | 50.2% | — |
| HumanEval+ | 72% | — |
| MBPP+ | 64.3% | — |

## Agentic & Tool Use

- Mixtral 8x22B: 22.8 (#119)
- Qwen2.5-Max: —

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| Cybench | 7.5% | — |

## Reasoning

- Mixtral 8x22B: 19.1 (#246)
- Qwen2.5-Max: 24.6 (#144)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1150 | 1360 |
| Epoch Capabilities Index | 122 | 132.54 |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 55.1% | — |
| LiveBench Data Analysis | — | 67.9% |
| ForecastBench | 56.3 | — |
| LiveBench | — | 62.3% |

## Math

- Mixtral 8x22B: 21.2 (#271)
- Qwen2.5-Max: 34.3 (#162)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1184 | 1369 |
| Omni-MATH | 16.3% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 24.2% | — |

## Knowledge

- Mixtral 8x22B: 15.1 (#288)
- Qwen2.5-Max: 34.8 (#181)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Expert | 1113 | 1337 |
| GPQA Diamond | 34.1% | — |
| MMLU-Pro | 46% | — |
| Confabulations | — | 21.8% |
| GPQA (HELM) | 33.4% | — |
| MMLU | 77.8% | — |

## Multilingual

- Mixtral 8x22B: 32.9 (#250)
- Qwen2.5-Max: 48.3 (#140)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1128 | 1352 |
| LMArena Chinese | 1116 | 1382 |
| LMArena French | 1165 | 1396 |
| LMArena German | 1141 | 1350 |
| LMArena Japanese | 1037 | 1299 |
| LMArena Korean | 1056 | 1304 |
| LMArena Russian | 1158 | 1353 |
| LMArena Spanish | 1150 | 1377 |

## Instruction Following

- Mixtral 8x22B: 57.9 (#261)
- Qwen2.5-Max: 71.5 (#147)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1147 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 72.4% | — |

## Long Context

- Mixtral 8x22B: 34.6 (#242)
- Qwen2.5-Max: 41.3 (#137)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1144 | 1358 |

## Writing & Preference

- Mixtral 8x22B: 37.1 (#257)
- Qwen2.5-Max: 55.6 (#141)

| Benchmark | Mixtral 8x22B | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1162 | 1367 |
| LMArena Creative Writing | 1141 | 1339 |
| LMArena Multi-Turn | 1130 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| WildBench | 71.1% | — |
| LiveBench Language | — | 56.3% |

## FAQ

### Is Mixtral 8x22B better than Qwen2.5-Max?

Qwen2.5-Max is the stronger model overall, scoring 40.1 to 26.7 on the Noometry Index.

### Is Mixtral 8x22B or Qwen2.5-Max better for coding?

Qwen2.5-Max scores higher on coding benchmarks: 41.2 versus 23.8 in the Noometry coding category.

### How many benchmarks do Mixtral 8x22B and Qwen2.5-Max share?

18 benchmarks have published results for both models. Mixtral 8x22B has 34 scored results on Noometry and Qwen2.5-Max has 27.