ORIGINAL WORKFLOW CREDIT GOES TO CREATOR @vslinx, link to their workflow as they are a hardworking creator! Try theirs before you try mine!❤️⬇️

https://civarchive.com/models/2237779/qwen-image-layered-simple-comfyui-workflow

I slightly condensed it for low vram!

• GGUF workflow for layering assets inside an image with Qwen-Image-Layered.

workflow includes lighting 8-step v2 lora for qwen but it admittedly doesnt work as well as the full sampling steps (20-25), use at your own discretion. i was able to get a few viable generations from it.

INSTALL MISSING CUSTOM NODES

ComfyUI-mxToolkit

ComfyUI-KJNodes

ComfyUI-Custom-Scripts

FILES

Model

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

VAE

https://huggingface.co/Comfy-Org/Qwen-Image-Layered_ComfyUI/tree/main/split_files/vae

Text encoder

https://huggingface.co/Comfy-Org/Qwen-Image_ComfyUI/blob/main/split_files/text_encoders/qwen_2.5_vl_7b_fp8_scaled.safetensors

LoRA (8-step v2)

https://huggingface.co/lightx2v/Qwen-Image-Lightning/tree/main

NOTE: if you can fit the Q4 or higher, i highly suggest using them instead of Q3 or lower. the quality drop makes Q3 and lower useless in my personal opinion.