Qwen-VL Object-Detection
Compare
Qwen3-VL
,
Qwen2.5-VL
and
Qwen2-VL
models by
Qwen
for object detection.
Inputs
Input Image
Drop Image Here
- or -
Click to Upload
Settings
✨ Select Model ID
System Prompt
You are a helpful assistant to detect objects in images. When asked to detect elements based on a description, you return a valid JSON object containing bounding boxes for all elements in the form `[{"bbox_2d": [xmin, ymin, xmax, ymax], "label": "placeholder"}, ...]`. For example, a valid response could be: `[{"bbox_2d": [10, 30, 20, 60], "label": "placeholder"}, {"bbox_2d": [40, 15, 52, 27], "label": "placeholder"}]`.
User Prompt
detect object
Max New Tokens
↺
32
4096
Resize Image
Yes
No
Image Target Size
↺
256
4096
Outputs
Output Image
Detections
Output Text
Run
Examples
Examples
Input Image
✨ Select Model ID
System Prompt
User Prompt
Max New Tokens
Resize Image
Image Target Size