# Mistral Large vs Qwen2.5 72B Instruct

> Mistral Large and Qwen2.5 72B Instruct score almost the same on the Noometry Index (31.6 vs 31.5), so choose on price, context window or the category you care about most.

- Canonical page: https://noometry.com/compare/mistral-large-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 33

## Summary

- They share 33 benchmarks with published results for both. Mistral Large scores higher in 4 categories and Qwen2.5 72B Instruct in 5 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in agentic & tool use, where Mistral Large leads 28.4 to 22.0.
- The biggest single-benchmark swing is BigCodeBench Complete: 38.3% for Mistral Large and 55.9% for Qwen2.5 72B Instruct.
- Qwen2.5 72B Instruct is cheaper at $1.40 / $5.60 per million input/output tokens, against $2 / $6 for Mistral Large.

## Snapshot

| | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | Mistral AI | Alibaba (Qwen) |
| Noometry Index | 31.6 | 31.5 |
| Rank | 258 | 259 |
| Context | 131K | 131K |
| Input $/M | $2 | $1.40 |
| Output $/M | $6 | $5.60 |
| Weights | Open | Open |

## Coding

- Mistral Large: 33.8 (#240)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| BigCodeBench Instruct | 30% | 45.8% |
| LMArena Coding | 1277 | 1293 |
| BigCodeBench Complete | 38.3% | 55.9% |
| SciCode | 36.2% | — |
| WeirdML | — | 16% |
| LiveBench Coding | 47.1% | — |
| ALE-Bench | 264.7 | — |
| HumanEval+ | 62.2% | — |
| MBPP+ | 59.5% | — |

## Agentic & Tool Use

- Mistral Large: 28.4 (#83)
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| Berkeley Function Calling Leaderboard | 38.4% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |

## Reasoning

- Mistral Large: 15.4 (#305)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1257 | 1271 |
| DTBench | 65.1% | 62.9% |
| LMCA | 16.7% | 13.4% |
| Epoch Capabilities Index | 128.52 | 129.01 |
| ForecastBench | 57.1 | 57.5 |
| SimpleBench | 22.5% | — |
| CritPt | 0% | — |
| LiveBench Reasoning | 43.5% | — |
| LiveBench Data Analysis | 50.1% | — |
| BIG-Bench Hard | — | 79.8% |
| HellaSwag | — | 84.8% |
| LiveBench | 48.4% | — |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- Mistral Large: 17.1 (#285)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| OTIS Mock AIME 2024-2025 | 8.5% | 8.1% |
| Omni-MATH | 28.1% | 33% |
| LMArena Math | 1262 | 1283 |
| MATH Level 5 | 50.3% | 63.2% |
| LiveBench Math | 42.5% | — |
| FrontierMath (Feb 2025 set) | 0.3% | — |

## Knowledge

- Mistral Large: 30.0 (#224)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| GPQA Diamond | 51.3% | 49.1% |
| MMLU-Pro | 59.9% | 63.1% |
| Confabulations | 21.4% | 19.1% |
| GPQA (HELM) | 43.5% | 42.6% |
| LMArena Expert | 1232 | 1245 |
| MMLU | 80% | 85.3% |
| Vectara Hallucination Rate | 4.5% | — |
| ARC (AI2) Challenge | — | 94.5% |
| TriviaQA | — | 71.9% |

## Multilingual

- Mistral Large: 40.1 (#214)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1237 | 1252 |
| LMArena Chinese | 1240 | 1272 |
| LMArena French | 1324 | 1279 |
| LMArena German | 1254 | 1234 |
| LMArena Japanese | 1187 | 1180 |
| LMArena Korean | 1202 | 1188 |
| LMArena Russian | 1257 | 1264 |
| LMArena Spanish | 1268 | 1255 |

## Instruction Following

- Mistral Large: 68.1 (#186)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| IFEval | 87.7% | 80.6% |
| LMArena Instruction Following | 1249 | 1254 |
| LiveBench Instruction Following | 67.9% | — |

## Long Context

- Mistral Large: 38.3 (#194)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1262 | 1282 |

## Writing & Preference

- Mistral Large: 40.9 (#237)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | Mistral Large | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1266 | 1269 |
| LMArena Creative Writing | 1243 | 1221 |
| WildBench | 80.1% | 80.2% |
| LMArena Multi-Turn | 1260 | 1272 |
| Short-Story Creative Writing | 69% | — |
| EQ-Bench Creative Writing | 985 | — |
| LiveBench Language | 39.4% | — |

## FAQ

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

Mistral Large and Qwen2.5 72B Instruct score almost the same on the Noometry Index (31.6 vs 31.5), so choose on price, context window or the category you care about most.

### Which is cheaper, Mistral Large or Qwen2.5 72B Instruct?

Qwen2.5 72B Instruct is cheaper. It lists at $1.40 per million input tokens and $5.60 per million output tokens; Mistral Large lists at $2 and $6.

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

Mistral Large scores higher on coding benchmarks: 33.8 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

Both accept 131K tokens.

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

33 benchmarks have published results for both models. Mistral Large has 51 scored results on Noometry and Qwen2.5 72B Instruct has 43.