# Mistral vs Qwen2.5 72B Instruct

> Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 29.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/mistral-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 22

## Summary

- They share 22 benchmarks with published results for both. Mistral scores higher in 2 categories and Qwen2.5 72B Instruct in 6 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in instruction following, where Qwen2.5 72B Instruct leads 65.5 to 52.6.
- The biggest single-benchmark swing is MMLU-Pro: 27.7% for Mistral and 63.1% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 29.9 | 31.9 |
| Rank | 303 | 267 |
| Context | — | 131K |
| Input $/M | — | $1.40 |
| Output $/M | — | $5.60 |
| Weights | Proprietary | Open |

## Coding

- Mistral: 33.8 (#250)
- Qwen2.5 72B Instruct: 33.2 (#260)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1162 | 1292 |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |

## Agentic & Tool Use

- Mistral: —
- Qwen2.5 72B Instruct: 22.1 (#133)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |

## Reasoning

- Mistral: 22.2 (#200)
- Qwen2.5 72B Instruct: 22.3 (#199)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1149 | 1271 |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- Mistral: 22.3 (#278)
- Qwen2.5 72B Instruct: 19.3 (#287)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| Omni-MATH | 7.2% | 33% |
| LMArena Math | 1180 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| MATH Level 5 | — | 63.2% |

## Knowledge

- Mistral: 16.6 (#288)
- Qwen2.5 72B Instruct: 27.0 (#253)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| MMLU-Pro | 27.7% | 63.1% |
| GPQA (HELM) | 30.3% | 42.6% |
| LMArena Expert | 1125 | 1245 |
| GPQA Diamond | — | 49.1% |
| Confabulations | — | 19.1% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multilingual

- Mistral: 32.8 (#254)
- Qwen2.5 72B Instruct: 41.0 (#213)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1129 | 1252 |
| LMArena Chinese | 1109 | 1272 |
| LMArena French | 1180 | 1280 |
| LMArena German | 1155 | 1234 |
| LMArena Japanese | 1013 | 1180 |
| LMArena Korean | 1032 | 1188 |
| LMArena Russian | 1168 | 1264 |
| LMArena Spanish | 1143 | 1256 |

## Instruction Following

- Mistral: 52.6 (#288)
- Qwen2.5 72B Instruct: 65.5 (#221)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 56.8% | 80.6% |
| LMArena Instruction Following | 1152 | 1254 |

## Long Context

- Mistral: 35.0 (#245)
- Qwen2.5 72B Instruct: 38.9 (#188)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1153 | 1282 |

## Writing & Preference

- Mistral: 37.0 (#260)
- Qwen2.5 72B Instruct: 46.7 (#215)

| Benchmark | Mistral | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1165 | 1269 |
| LMArena Creative Writing | 1158 | 1221 |
| WildBench | 66% | 80.2% |
| LMArena Multi-Turn | 1147 | 1272 |

## FAQ

### Is Mistral better than Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is the stronger model overall, scoring 31.9 to 29.9 on the Noometry Index.

### Is Mistral or Qwen2.5 72B Instruct better for coding?

They score almost the same on coding (33.8 vs 33.2); test both on your own repository before choosing.

### How many benchmarks do Mistral and Qwen2.5 72B Instruct share?

22 benchmarks have published results for both models. Mistral has 22 scored results on Noometry and Qwen2.5 72B Instruct has 43.