# Qwen-14B vs Qwen Max

> Qwen Max is the stronger model overall, scoring 34.7 to 31.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen-14b-vs-qwen-max
- Last updated: 2026-10-10
- Shared benchmarks: 10

## Summary

- They share 10 benchmarks with published results for both. Qwen-14B scores higher in 2 categories and Qwen Max in 5 categories; 6 gaps are clear of the uncertainty.
- The widest gap is in writing & preference, where Qwen Max leads 47.8 to 27.6.
- Qwen-14B has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen-14B | Qwen Max |
|---|---|---|
| Provider | Alibaba (Qwen) | Alibaba (Qwen) |
| Noometry Index | 31.4 | 34.7 |
| Rank | 275 | 230 |
| Context | — | 33K |
| Input $/M | — | $1.60 |
| Output $/M | — | $6.40 |
| Weights | Open | Proprietary |

## Coding

- Qwen-14B: 31.2 (#288)
- Qwen Max: 30.7 (#292)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Coding | 1071 | 1288 |
| Aider Polyglot | — | 21.8% |

## Reasoning

- Qwen-14B: 19.6 (#257)
- Qwen Max: 25.1 (#151)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Hard Prompts | 1027 | 1269 |
| BIG-Bench Hard | 55% | — |
| Epoch Capabilities Index | 113.03 | — |
| LAMBADA | 71.1% | — |
| PIQA | 79.9% | — |

## Math

- Qwen-14B: 31.2 (#227)
- Qwen Max: 22.3 (#276)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Math | 1068 | 1275 |
| OTIS Mock AIME 2024-2025 | — | 16.1% |
| MATH Level 5 | — | 67.2% |
| FrontierMath (Feb 2025 set) | — | 1% |
| GSM8K | 61.3% | — |

## Knowledge

- Qwen-14B: —
- Qwen Max: 30.3 (#228)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| GPQA Diamond | — | 56.1% |
| LMArena Expert | — | 1248 |
| ARC (AI2) Challenge | 84.4% | — |
| BoolQ | 86.2% | — |
| MMLU | 66.3% | — |

## Multilingual

- Qwen-14B: 27.5 (#275)
- Qwen Max: 41.8 (#202)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Non-English | 1041 | 1263 |
| LMArena Chinese | 1077 | 1254 |
| LMArena French | — | 1330 |
| LMArena German | — | 1254 |
| LMArena Japanese | — | 1205 |
| LMArena Korean | — | 1142 |
| LMArena Russian | — | 1274 |
| LMArena Spanish | — | 1290 |

## Instruction Following

- Qwen-14B: 52.4 (#289)
- Qwen Max: 66.5 (#208)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Instruction Following | 1031 | 1262 |

## Long Context

- Qwen-14B: 31.3 (#280)
- Qwen Max: 39.4 (#180)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Longer Query | 1028 | 1288 |
| Fiction.LiveBench | — | 66.7% |

## Writing & Preference

- Qwen-14B: 27.6 (#299)
- Qwen Max: 47.8 (#205)

| Benchmark | Qwen-14B | Qwen Max |
|---|---|---|
| LMArena Text | 1051 | 1282 |
| LMArena Creative Writing | 1028 | 1248 |
| LMArena Multi-Turn | 1022 | 1277 |

## FAQ

### Is Qwen-14B better than Qwen Max?

Qwen Max is the stronger model overall, scoring 34.7 to 31.4 on the Noometry Index.

### Is Qwen-14B or Qwen Max better for coding?

They score almost the same on coding (31.2 vs 30.7); test both on your own repository before choosing.

### How many benchmarks do Qwen-14B and Qwen Max share?

10 benchmarks have published results for both models. Qwen-14B has 18 scored results on Noometry and Qwen Max has 23.