Support for DiffSynth-Studio's Eligen-v2 for Qwen Image

[LuLmaster69]

Feature Idea

Now that ComfyUI supports the controlnets for Qwen Image from DiffSynth-Studio, I think it would it timely to have support for their Eligen-v2 lora that allows for prompt instructions to be written on the image with used of color masked areas to direct the model on where to generate objects on the picture.

Please see the lora model and examples here https://modelscope.cn/models/DiffSynth-Studio/Qwen-Image-EliGen-V2

Existing Solutions

The Eilgen-v2 lora does not work on ComfyUI as of this date and there are no custom nodes that address this.

Other

No response

[VitoChenLY]

Are there any adaptation plans

[krisheetu]

Hey @VitoChenLY @LuLmaster69 , check this custom node, it is using diffsynth pipeline but kinda works in comfyUI:

https://github.com/AIFSH/QwenImage-Diffsynth ,and yes, you can test version 2 here as well.

[geroldmeisinger]

dup of #9993
see this PR #10473