this is the setup i am using to test the different version of qwen image edit, and i thought i would share::

setup with bypasser

has an sdxl that qwen image feed thru for fun

has an extra image you can turn on to load two images as input

setting up gguf

https://huggingface.co/QuantStack/Qwen-Image-Edit-GGUF

qquf i am using for text encoder

https://huggingface.co/QuantStack/Qwen-Image-Edit-GGUF/tree/main/mmproj

https://huggingface.co/unsloth/Qwen2.5-VL-7B-Instruct-GGUF/tree/main

prompt manager

https://github.com/ComfyAssets/ComfyUI_PromptManager

lora loader

https://github.com/willmiao/ComfyUI-Lora-Manager

lora loader can only have i image save with embed workflow-- i have two one for a extra run thru sdxl for fun : ) you can turn it off

sampler

https://github.com/ClownsharkBatwing/RES4LYF

exponential/res_2s :::: beta57

-- comfyui needs to be update to the latest for this workflow-- using the bat

-- not a pro just good at reading the labels on the tin and following directions

-- if you have issues i might be able to help