lightx2v

/

Qwen-Image-Lightning

like 808

Hi, it may be related with comfyUI version. this link may help you.

Same for me images are blurry. im using ComfyUI v0.3.52 and my GPU is L40s. Im using the FP8 version of the Qwen-Image-Edit model.
i've tried the Qwen-Image-Edit-Lightning-4steps-V1.0-bf16.safetensors and
Qwen-Image-Edit-Lightning-4steps-V1.0.safetensors but both are blurry.

However Qwen-Image-Lightning-4steps-V1.0.safetensors is working very good with Qwen-Image-Edit 🤷‍♂️

EDIT: The issue seems to occur only when doing text2img (I've tried the same workflow with and without an input image)

Hi, it may be related with comfyUI version. this link may help you.

hi, bro. How dose this lightning been trained? Can u share some tips and dataset ?

according to this subreddit txt2img doesn't work with the qwen-image-edit lora: https://github.com/ModelTC/Qwen-Image-Lightning/issues/26#issuecomment-3218541520