Qwen-Image-2.1-Object-Mover-Bbox-turbo

Qwen-Image-2.1-Object-Mover-Bbox-turbo is a Turbo-optimized LoRA adapter for Qwen-Image-2.1, designed for object movement using bounding box guidance. The adapter moves an object highlighted by a red bounding box to the location indicated by another red bounding box while attempting to preserve the object's appearance, surrounding textures, lighting, shadows, perspective, and overall image consistency.

The adapter is optimized for fast Turbo inference and can also be used with standard inference workflows.

This is an experimental model. Results may vary depending on the input image, object, bounding boxes, target location, and surrounding context.

Model Details

• Base Model: Qwen/Qwen-Image-2.1
• Adapter: Qwen-Image-2.1-Object-Mover-Bbox-turbo
• Model Type: LoRA / Adapter
• Model Status: Experimental
• Inference: Turbo / Standard
• Created by: prithivMLmods

Training Specifications

Parameter	Configuration
Dataset	45 pairs of high-quality images with bounding box annotations and manually manipulated resultant images
Save Precision	BF16
Learning Rate	1e-4
Optimizer	AdamW
Network Dimension (Rank)	16
Total Steps	4000
Trigger Prompt	Move the object highlighted in the red box to the location indicated by the other red box in the scene.

Download

Download the model files from the Files & versions tab:

Download Qwen-Image-2.1-Object-Mover-Bbox-turbo

Usage

Load the LoRA adapter with Qwen-Image-2.1 and provide an image containing two red bounding boxes: one highlighting the object to be moved and another indicating its target location.

Use the trigger prompt:

Move the object highlighted in the red box to the location indicated by the other red box in the scene.

For best results, use clear bounding boxes around both the source object and target location.

Inference

This adapter is optimized for fast Turbo inference and can also be used with standard inference workflows.

For Turbo inference, follow the recommended inference configuration for Qwen-Image-2.1 and apply the adapter using the appropriate Turbo workflow.

Limitations

This is an experimental release and may produce artifacts or inconsistencies in challenging cases, including:

• Complex backgrounds
• Large or overlapping objects
• Fine structures and textures
• Reflections and transparent objects
• Difficult lighting and perspective conditions
• Ambiguous or poorly positioned bounding boxes
• Significant changes in object scale or perspective

Results can vary depending on the input image, object, bounding box placement, and target location.

License

Please refer to the license terms of the base model, Qwen-Image-2.1, and ensure compliance with its terms when using or redistributing this adapter.

Acknowledgements

This adapter was created by prithivMLmods and is built for use with Qwen-Image-2.1.