# Qwen2.5-VL 72B Instruct vs Sonar

> Sonar is the stronger model overall, scoring 37.8 to 29.4 on the Noometry Index.

- Canonical page: https://noometry.com/compare/qwen2-5-vl-72b-instruct-vs-sonar
- Last updated: 2026-10-10
- Shared benchmarks: 0

## Summary

- Sonar is cheaper at $1 / $1 per million input/output tokens, against $2.80 / $8.40 for Qwen2.5-VL 72B Instruct.
- Qwen2.5-VL 72B Instruct accepts more context: 131K tokens versus 128K.
- Qwen2.5-VL 72B Instruct has downloadable open weights; the other is API-only.

## Snapshot

| | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| Provider | Alibaba (Qwen) | Perplexity |
| Noometry Index | 29.4 | 37.8 |
| Rank | 298 | 184 |
| Context | 131K | 128K |
| Input $/M | $2.80 | $1 |
| Output $/M | $8.40 | $1 |
| Weights | Open | Proprietary |

## Coding

- Qwen2.5-VL 72B Instruct: —
- Sonar: 35.1 (#217)

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| LiveBench Coding | — | 35.1% |

## Agentic & Tool Use

- Qwen2.5-VL 72B Instruct: 18.4 (#136)
- Sonar: —

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| OSWorld | 5% | — |

## Reasoning

- Qwen2.5-VL 72B Instruct: 20.0 (#228)
- Sonar: 20.3 (#221)

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| Kagi LLM Benchmark | 36% | — |
| LiveBench Reasoning | — | 46.3% |
| LiveBench Data Analysis | — | 37.9% |
| LiveBench | — | 46.9% |

## Math

- Qwen2.5-VL 72B Instruct: —
- Sonar: 31.3 (#197)

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| LiveBench Math | — | 41.6% |

## Multimodal

- Qwen2.5-VL 72B Instruct: 32.9 (#92)
- Sonar: —

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| LMArena Vision | 1107 | — |
| Video-MME | 73.5% | — |
| GeoBench | 62% | — |
| SpatialViz-Bench | 33.3% | — |

## Instruction Following

- Qwen2.5-VL 72B Instruct: —
- Sonar: 71.5 (#146)

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| LiveBench Instruction Following | — | 76.2% |

## Writing & Preference

- Qwen2.5-VL 72B Instruct: —
- Sonar: 52.6 (#166)

| Benchmark | Qwen2.5-VL 72B Instruct | Sonar |
|---|---|---|
| LiveBench Language | — | 44.1% |

## FAQ

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

Sonar is the stronger model overall, scoring 37.8 to 29.4 on the Noometry Index.

### Which is cheaper, Qwen2.5-VL 72B Instruct or Sonar?

Sonar is cheaper. It lists at $1 per million input tokens and $1 per million output tokens; Qwen2.5-VL 72B Instruct lists at $2.80 and $8.40.

### Which has the bigger context window?

Qwen2.5-VL 72B Instruct does, with 131K tokens against 128K.

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

0 benchmarks have published results for both models. Qwen2.5-VL 72B Instruct has 6 scored results on Noometry and Sonar has 7.