# MiniMax-M2 vs Qwen2.5 72B Instruct

> MiniMax-M2 is the stronger model overall, scoring 36.9 to 31.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/minimax-m2-vs-qwen2-5-72b-instruct
- Last updated: 2026-10-10
- Shared benchmarks: 15

## Summary

- They share 15 benchmarks with published results for both. MiniMax-M2 scores higher in 8 categories and Qwen2.5 72B Instruct in 1 category; 9 gaps are clear of the uncertainty.
- The widest gap is in math, where MiniMax-M2 leads 34.6 to 18.1.
- MiniMax-M2 is cheaper at $0.30 / $1.20 per million input/output tokens, against $1.40 / $5.60 for Qwen2.5 72B Instruct.
- MiniMax-M2 accepts more context: 205K tokens versus 131K.

## Snapshot

| | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| Provider | MiniMax | Alibaba (Qwen) |
| Noometry Index | 36.9 | 31.5 |
| Rank | 201 | 259 |
| Context | 205K | 131K |
| Input $/M | $0.30 | $1.40 |
| Output $/M | $1.20 | $5.60 |
| Weights | Open | Open |

## Coding

- MiniMax-M2: 39.1 (#151)
- Qwen2.5 72B Instruct: 32.8 (#256)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Coding | 1370 | 1293 |
| SWE-bench Verified (bash only) | 61% | — |
| LMArena WebDev | 1297 | — |
| WeirdML | — | 16% |
| BigCodeBench Instruct | — | 45.8% |
| BigCodeBench Complete | — | 55.9% |

## Agentic & Tool Use

- MiniMax-M2: 24.9 (#102)
- Qwen2.5 72B Instruct: 22.0 (#124)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| Terminal-Bench | 30% | — |
| TheAgentCompany | — | 5.7% |
| BALROG | — | 16.2% |
| METR Time Horizons | — | 35.8% |
| Vending-Bench 2 | 160.6 | — |

## Reasoning

- MiniMax-M2: 18.6 (#256)
- Qwen2.5 72B Instruct: 21.4 (#197)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Hard Prompts | 1357 | 1271 |
| Kagi LLM Benchmark | 57.8% | — |
| NYT Connections (extended) | 14.8% | — |
| DTBench | — | 62.9% |
| LMCA | — | 13.4% |
| BIG-Bench Hard | — | 79.8% |
| Epoch Capabilities Index | — | 129.01 |
| ForecastBench | — | 57.5 |
| HellaSwag | — | 84.8% |
| PIQA | — | 82.6% |
| WinoGrande | — | 82.3% |

## Math

- MiniMax-M2: 34.6 (#158)
- Qwen2.5 72B Instruct: 18.1 (#281)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Math | 1350 | 1283 |
| OTIS Mock AIME 2024-2025 | — | 8.1% |
| Omni-MATH | — | 33% |
| MATH Level 5 | — | 63.2% |

## Knowledge

- MiniMax-M2: 36.3 (#160)
- Qwen2.5 72B Instruct: 26.9 (#247)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Expert | 1336 | 1245 |
| GPQA Diamond | — | 49.1% |
| MMLU-Pro | — | 63.1% |
| Confabulations | — | 19.1% |
| GPQA (HELM) | — | 42.6% |
| ARC (AI2) Challenge | — | 94.5% |
| MMLU | — | 85.3% |
| TriviaQA | — | 71.9% |

## Multilingual

- MiniMax-M2: 45.6 (#165)
- Qwen2.5 72B Instruct: 41.2 (#208)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Non-English | 1314 | 1252 |
| LMArena Chinese | 1367 | 1272 |
| LMArena French | 1336 | 1279 |
| LMArena German | 1366 | 1234 |
| LMArena Russian | 1332 | 1264 |
| LMArena Spanish | 1326 | 1255 |
| LMArena Japanese | — | 1180 |
| LMArena Korean | — | 1188 |

## Instruction Following

- MiniMax-M2: 70.4 (#160)
- Qwen2.5 72B Instruct: 65.7 (#216)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Instruction Following | 1329 | 1254 |
| IFEval | — | 80.6% |

## Long Context

- MiniMax-M2: 40.5 (#148)
- Qwen2.5 72B Instruct: 38.9 (#183)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Longer Query | 1332 | 1282 |

## Writing & Preference

- MiniMax-M2: 53.3 (#157)
- Qwen2.5 72B Instruct: 46.9 (#210)

| Benchmark | MiniMax-M2 | Qwen2.5 72B Instruct |
|---|---|---|
| LMArena Text | 1340 | 1269 |
| LMArena Creative Writing | 1288 | 1221 |
| LMArena Multi-Turn | 1362 | 1272 |
| WildBench | — | 80.2% |

## FAQ

### Is MiniMax-M2 better than Qwen2.5 72B Instruct?

MiniMax-M2 is the stronger model overall, scoring 36.9 to 31.5 on the Noometry Index.

### Which is cheaper, MiniMax-M2 or Qwen2.5 72B Instruct?

MiniMax-M2 is cheaper. It lists at $0.30 per million input tokens and $1.20 per million output tokens; Qwen2.5 72B Instruct lists at $1.40 and $5.60.

### Is MiniMax-M2 or Qwen2.5 72B Instruct better for coding?

MiniMax-M2 scores higher on coding benchmarks: 39.1 versus 32.8 in the Noometry coding category.

### Which has the bigger context window?

MiniMax-M2 does, with 205K tokens against 131K.

### How many benchmarks do MiniMax-M2 and Qwen2.5 72B Instruct share?

15 benchmarks have published results for both models. MiniMax-M2 has 21 scored results on Noometry and Qwen2.5 72B Instruct has 43.