User Manual

1.This workflow uses Qwen-Image-Edit-Rapid-AIO for image editing, supporting single-image, dual-image and triple-image editing, or generating storyboards from a single image.

2.For storyboard usage, enable the toggle and manually connect the prompts. Storyboards can be generated via Qwen3VL inference by simply modifying the number of generated storyboard frames and the plot. When using the Qwen3VL model, the 8B parameter version is recommended for better performance. Gemini or Gork can also be used to generate storyboard prompts for optimal results.

3.For multi-image usage, enable the toggle and select the relevant option as needed.

4.Enable the HD and face restoration features according to your requirements.

5.The latest version of the Qwen-Image-Edit-Rapid-AIO model is used for editing. This model is highly user-friendly, yet there are noticeable differences between different versions; you can download and use the appropriate version based on your specific needs.

6.Two LoRAs are included and can be enabled as required. F2P enhances facial consistency significantly but may reduce the diversity of facial expressions and available angles. Tests have shown that the NSFW feature of MCNL is compatible with the AIO model, providing better support for NSFW content generation.

7.The Parameter Selection Area allows for resolution control.

8.Adjust all other parameters according to your specific needs.

This workflow is available for free online testing—download it for local use only if it meets your needs.

Access address:https://www.runninghub.ai/post/2017931773761556482/?inviteCode=rh-v1443

DeepwhiteAI: https://civarchive.com/user/DeepWhiteAI

Acknowledgements:

Phr00t: https://huggingface.co/Phr00t/Qwen-Image-Edit-Rapid-AIO