# Qwen2.5 72B Instruct vs Qwen2.5-Max

> Qwen2.5-Max is the stronger model overall, scoring 40.7 to 31.9 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen2-5-72b-instruct-vs-qwen2-5-max
- Last updated: 2026-10-12
- Shared benchmarks: 19

## Summary

- They share 19 benchmarks with published results for both. Qwen2.5 72B Instruct scores higher in 0 categories and Qwen2.5-Max in 8 categories; 8 gaps are clear of the uncertainty.
- The widest gap is in math, where Qwen2.5-Max leads 36.9 to 19.3.
- Qwen2.5 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 31.9 | 40.7 |
| Rank | 267 | 146 |
| Context | 131K | — |
| Input $/M | $1.40 | — |
| Output $/M | $5.60 | — |
| Weights | Open | Proprietary |

## Coding

- Qwen2.5 72B Instruct: 33.2 (#260)
- Qwen2.5-Max: 41.8 (#117)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Coding | 1292 | 1359 |
| WeirdML | 16% | — |
| BigCodeBench Instruct | 45.8% | — |
| LiveBench Coding | — | 64.4% |
| BigCodeBench Complete | 55.9% | — |

## Agentic & Tool Use

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

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

## Reasoning

- Qwen2.5 72B Instruct: 22.3 (#199)
- Qwen2.5-Max: 25.6 (#147)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Hard Prompts | 1271 | 1360 |
| Epoch Capabilities Index | 129 | 132.53 |
| LiveBench Reasoning | — | 51.4% |
| DTBench | 62.9% | — |
| LiveBench Data Analysis | — | 67.9% |
| LMCA | 13.4% | — |
| BIG-Bench Hard | 79.8% | — |
| ForecastBench | 57.5 | — |
| HellaSwag | 84.8% | — |
| LiveBench | — | 62.3% |
| PIQA | 82.6% | — |
| WinoGrande | 82.3% | — |

## Math

- Qwen2.5 72B Instruct: 19.3 (#287)
- Qwen2.5-Max: 36.9 (#162)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Math | 1283 | 1369 |
| OTIS Mock AIME 2024-2025 | 8.1% | — |
| Omni-MATH | 33% | — |
| LiveBench Math | — | 58.4% |
| MATH Level 5 | 63.2% | — |

## Knowledge

- Qwen2.5 72B Instruct: 27.0 (#253)
- Qwen2.5-Max: 35.3 (#186)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| Confabulations | 19.1% | 21.8% |
| LMArena Expert | 1245 | 1337 |
| GPQA Diamond | 49.1% | — |
| MMLU-Pro | 63.1% | — |
| GPQA (HELM) | 42.6% | — |
| ARC (AI2) Challenge | 94.5% | — |
| MMLU | 85.3% | — |
| TriviaQA | 71.9% | — |

## Multilingual

- Qwen2.5 72B Instruct: 41.0 (#213)
- Qwen2.5-Max: 48.1 (#146)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Non-English | 1252 | 1352 |
| LMArena Chinese | 1272 | 1382 |
| LMArena French | 1280 | 1396 |
| LMArena German | 1234 | 1350 |
| LMArena Japanese | 1180 | 1300 |
| LMArena Korean | 1188 | 1304 |
| LMArena Russian | 1264 | 1353 |
| LMArena Spanish | 1256 | 1377 |

## Instruction Following

- Qwen2.5 72B Instruct: 65.5 (#221)
- Qwen2.5-Max: 71.3 (#152)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Instruction Following | 1254 | 1335 |
| LiveBench Instruction Following | — | 75.3% |
| IFEval | 80.6% | — |

## Long Context

- Qwen2.5 72B Instruct: 38.9 (#188)
- Qwen2.5-Max: 41.4 (#142)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Longer Query | 1282 | 1358 |

## Writing & Preference

- Qwen2.5 72B Instruct: 46.7 (#215)
- Qwen2.5-Max: 55.4 (#146)

| Benchmark | Qwen2.5 72B Instruct | Qwen2.5-Max |
|---|---|---|
| LMArena Text | 1269 | 1367 |
| LMArena Creative Writing | 1221 | 1339 |
| LMArena Multi-Turn | 1272 | 1364 |
| Short-Story Creative Writing | — | 72.9% |
| WildBench | 80.2% | — |
| LiveBench Language | — | 56.3% |

## FAQ

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

Qwen2.5-Max is the stronger model overall, scoring 40.7 to 31.9 on the Noometry Index.

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

Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 33.2 in the Noometry coding category.

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

19 benchmarks have published results for both models. Qwen2.5 72B Instruct has 43 scored results on Noometry and Qwen2.5-Max has 27.