# Qwen2.5-Max vs Sonar

> Qwen2.5-Max is the stronger model overall, scoring 40.7 to 38.5 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen2-5-max-vs-sonar
- Last updated: 2026-10-12
- Shared benchmarks: 7

## Summary

- They share 7 benchmarks with published results for both. Qwen2.5-Max scores higher in 4 categories and Sonar in 1 category; 4 gaps are clear of the uncertainty.
- The widest gap is in coding, where Qwen2.5-Max leads 41.8 to 35.7.
- The biggest single-benchmark swing is LiveBench Data Analysis: 67.9% for Qwen2.5-Max and 37.9% for Sonar.

## Snapshot

| | Qwen2.5-Max | Sonar |
|---|---|---|
| Provider | Alibaba (Qwen) | Perplexity |
| Noometry Index | 40.7 | 38.5 |
| Rank | 146 | 187 |
| Context | — | 128K |
| Input $/M | — | $1 |
| Output $/M | — | $1 |
| Weights | Proprietary | Proprietary |

## Coding

- Qwen2.5-Max: 41.8 (#117)
- Sonar: 35.7 (#221)

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LiveBench Coding | 64.4% | 35.1% |
| LMArena Coding | 1359 | — |

## Reasoning

- Qwen2.5-Max: 25.6 (#147)
- Sonar: 21.1 (#227)

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LiveBench Reasoning | 51.4% | 46.3% |
| LiveBench Data Analysis | 67.9% | 37.9% |
| LiveBench | 62.3% | 46.9% |
| LMArena Hard Prompts | 1360 | — |
| Epoch Capabilities Index | 132.53 | — |

## Math

- Qwen2.5-Max: 36.9 (#162)
- Sonar: 33.7 (#200)

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LiveBench Math | 58.4% | 41.6% |
| LMArena Math | 1369 | — |

## Knowledge

- Qwen2.5-Max: 35.3 (#186)
- Sonar: —

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| Confabulations | 21.8% | — |
| LMArena Expert | 1337 | — |

## Multilingual

- Qwen2.5-Max: 48.1 (#146)
- Sonar: —

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LMArena Non-English | 1352 | — |
| LMArena Chinese | 1382 | — |
| LMArena French | 1396 | — |
| LMArena German | 1350 | — |
| LMArena Japanese | 1300 | — |
| LMArena Korean | 1304 | — |
| LMArena Russian | 1353 | — |
| LMArena Spanish | 1377 | — |

## Instruction Following

- Qwen2.5-Max: 71.3 (#152)
- Sonar: 71.4 (#150)

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LiveBench Instruction Following | 75.3% | 76.2% |
| LMArena Instruction Following | 1335 | — |

## Long Context

- Qwen2.5-Max: 41.4 (#142)
- Sonar: —

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LMArena Longer Query | 1358 | — |

## Writing & Preference

- Qwen2.5-Max: 55.4 (#146)
- Sonar: 52.6 (#167)

| Benchmark | Qwen2.5-Max | Sonar |
|---|---|---|
| LiveBench Language | 56.3% | 44.1% |
| LMArena Text | 1367 | — |
| LMArena Creative Writing | 1339 | — |
| Short-Story Creative Writing | 72.9% | — |
| LMArena Multi-Turn | 1364 | — |

## FAQ

### Is Qwen2.5-Max better than Sonar?

Qwen2.5-Max is the stronger model overall, scoring 40.7 to 38.5 on the Noometry Index.

### Is Qwen2.5-Max or Sonar better for coding?

Qwen2.5-Max scores higher on coding benchmarks: 41.8 versus 35.7 in the Noometry coding category.

### How many benchmarks do Qwen2.5-Max and Sonar share?

7 benchmarks have published results for both models. Qwen2.5-Max has 27 scored results on Noometry and Sonar has 7.