Qwen3-VL

Run Qwen3-VL on an image.

Deprecated

Use the unified Qwen-VL block (roboflow_core/qwen_vlm@v1), which exposes Qwen 3 VL alongside other Qwen variants and the OpenRouter passthrough.

This workflow block runs Qwen3-VL-a vision language model that accepts an image and an optional text prompt-and returns a text answer based on a conversation template.

Type identifier

Use the following identifier in step "type" field: roboflow_core/qwen3vl@v1 to add the block as a step in your workflow.

Properties

Name	Type	Description	Refs
name	str	Enter a unique identifier for this step..	❌
prompt	str	Optional text prompt to provide additional context to Qwen3-VL. Otherwise it will just be a default one, which may affect the desired model behavior..	❌
model_version	str	The Qwen3-VL model to be used for inference..	✅
system_prompt	str	Optional system prompt to provide additional context to Qwen3-VL..	❌

The Refs column marks possibility to parametrise the property with dynamic values available in workflow runtime. See Bindings for more info.

Runtime compatibility

hard - runtime self_hosted_cpu; execution local : Requires a GPU; run_locally() loads a model that needs CUDA.

Input and Output Bindings

The available connections depend on its binding kinds. Check what binding kinds Qwen3-VL in version v1 has.

Input and output bindings

• input

  • images (image): The image to infer on..

  • model_version (roboflow_model_id): The Qwen3-VL model to be used for inference..

• output

  • parsed_output (dictionary): Dictionary.

Example JSON definition

{
	    "name": "<your_step_name_here>",
	    "type": "roboflow_core/qwen3vl@v1",
	    "images": "$inputs.image",
	    "prompt": "What is in this image?",
	    "model_version": "qwen3vl-2b-instruct",
	    "system_prompt": "You are a helpful assistant."
	}