# Qwen Max vs Qwen2-72B

> Qwen Max is the stronger model overall, scoring 34.4 to 29.5 on the Noometry Index.

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

## Summary

- They share 19 benchmarks with published results for both. Qwen Max scores higher in 7 categories and Qwen2-72B in 1 category; 8 gaps are clear of the uncertainty.
- The widest gap is in knowledge, where Qwen Max leads 30.3 to 21.2.
- The biggest single-benchmark swing is MATH Level 5: 67.2% for Qwen Max and 39.1% for Qwen2-72B.
- Qwen2-72B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen Max | Qwen2-72B |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 34.4 | 29.5 |
| Rank | 226 | 294 |
| Context | 33K | — |
| Input $/M | $1.60 | — |
| Output $/M | $6.40 | — |
| Weights | Proprietary | Open |

## Coding

- Qwen Max: 30.3 (#288)
- Qwen2-72B: 28.7 (#303)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Coding | 1288 | 1196 |
| Aider Polyglot | 21.8% | — |
| WeirdML | — | 11.3% |
| BigCodeBench Instruct | — | 38.5% |
| BigCodeBench Complete | — | 54% |

## Agentic & Tool Use

- Qwen Max: —
- Qwen2-72B: 16.9 (#138)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| TheAgentCompany | — | 1.1% |
| METR Time Horizons | — | 29.9% |

## Reasoning

- Qwen Max: 24.1 (#154)
- Qwen2-72B: 22.2 (#184)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Hard Prompts | 1270 | 1191 |
| Epoch Capabilities Index | — | 125.27 |

## Math

- Qwen Max: 21.0 (#272)
- Qwen2-72B: 27.9 (#232)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Math | 1275 | 1235 |
| MATH Level 5 | 67.2% | 39.1% |
| OTIS Mock AIME 2024-2025 | 16.1% | — |
| FrontierMath (Feb 2025 set) | 1% | — |

## Knowledge

- Qwen Max: 30.3 (#222)
- Qwen2-72B: 21.2 (#270)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| GPQA Diamond | 56.1% | 40.8% |
| LMArena Expert | 1248 | 1171 |
| MMLU | — | 82.4% |

## Multilingual

- Qwen Max: 42.0 (#197)
- Qwen2-72B: 36.0 (#239)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Non-English | 1263 | 1176 |
| LMArena Chinese | 1254 | 1240 |
| LMArena French | 1329 | 1169 |
| LMArena German | 1254 | 1151 |
| LMArena Japanese | 1205 | 1111 |
| LMArena Korean | 1141 | 1083 |
| LMArena Russian | 1274 | 1169 |
| LMArena Spanish | 1290 | 1168 |

## Instruction Following

- Qwen Max: 66.7 (#203)
- Qwen2-72B: 62.0 (#236)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Instruction Following | 1262 | 1181 |

## Long Context

- Qwen Max: 39.4 (#175)
- Qwen2-72B: 36.1 (#230)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Longer Query | 1289 | 1193 |
| Fiction.LiveBench | 66.7% | — |

## Writing & Preference

- Qwen Max: 48.0 (#200)
- Qwen2-72B: 41.0 (#236)

| Benchmark | Qwen Max | Qwen2-72B |
|---|---|---|
| LMArena Text | 1282 | 1203 |
| LMArena Creative Writing | 1248 | 1181 |
| LMArena Multi-Turn | 1277 | 1195 |

## FAQ

### Is Qwen Max better than Qwen2-72B?

Qwen Max is the stronger model overall, scoring 34.4 to 29.5 on the Noometry Index.

### Is Qwen Max or Qwen2-72B better for coding?

Qwen Max scores higher on coding benchmarks: 30.3 versus 28.7 in the Noometry coding category.

### How many benchmarks do Qwen Max and Qwen2-72B share?

19 benchmarks have published results for both models. Qwen Max has 23 scored results on Noometry and Qwen2-72B has 26.