← all models

Qwen3 VL 32B Instruct

qwen

openrouter
Parameters 32B openrouter api · from the model id
Context 128K 131072 tokens
Input $0.10 OpenRouter
Output $0.42 OpenRouter
Cache read — OpenRouter

Accepts

text image

Capabilities

frequency_penalty logprobs max_tokens presence_penalty response_format seed stop structured_outputs temperature tool_choice tools top_k top_logprobs top_p

Description

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...

The OpenRouter API truncates descriptions — read the full text at the source

Also known as

Every id and name this model has been seen under — for matching it against another catalogue.

openrouter qwen/qwen3-vl-32b-instruct huggingface Qwen/Qwen3-VL-32B-Instruct name Qwen: Qwen3 VL 32B Instruct display name Qwen3 VL 32B Instruct

Identifiers

Vendor
qwen
OpenRouter
qwen/qwen3-vl-32b-instruct
Hugging Face
Qwen/Qwen3-VL-32B-Instruct
Released
2025-10-23T14:55:32.000Z
First seen
59 d ago
Last crawled
40 d ago