---
title: "Qwen: Qwen3 VL 32B Instruct — price, capability and what beats it · Undominated.ai"
canonical: https://undominated.ai/models/qwen__qwen3-vl-32b-instruct/
description: "Qwen: Qwen3 VL 32B Instruct: $0.104/M in, $0.416/M out. Independent capability scores, context-tier pricing, and the models that are both better and cheaper."
---

# Qwen: Qwen3 VL 32B Instruct — price, capability and what beats it · Undominated.ai

> Qwen: Qwen3 VL 32B Instruct: $0.104/M in, $0.416/M out. Independent capability scores, context-tier pricing, and the models that are both better and cheaper.

[Leaderboard](/) / Qwen

# Qwen3 VL 32B Instruct

Qwen · released 2025-10-23

 $0.182 per million tokens, balanced Balanced Summarise Chat Code gen Agentic

## No independent quality score.

No benchmark we track has measured this model. That is not the same as measuring it and finding it wanting — we simply cannot rank it, so we do not.

### Every price dimension

| Input | $0.104 /M |
| --- | --- |
| Output | $0.416 /M |

### Independent scores

No independent benchmark has measured this model.

### Capability

| Context window | 131K |
| --- | --- |
| Max output | 33K |
| Input modes | text, image |
| Tool use | yes |
| Reasoning | no |

### Provenance

| Price source | openrouter.ai |
| --- | --- |
| Fetched | 2026-08-24 |
| Quality data | unrated |

Qwen3-VL-32B-Instruct is a large-scale multimodal vision-language model designed for high-precision understanding and reasoning across text, images, and video. With 32 billion parameters, it combines deep visual perception with advanced text...