Qwen LoRA Consistent Character Creation Workflow
My modified version of the standard Qwen workflow that supports the creation of a LoRA dataset by generating multiple images from one reference image and allowing you to end up with a dataset that can be used to train different LoRAs of characters.
Download the specific designed Prompts from my Kofi page for free
Workflow Information
This workflow uses the following nodes:
- CFGNorm
- CLIPLoader
- EmptySD3LatentImage
- KSampler
- LoadImage
- ModelSamplingAuraFlow
- SaveImage
- Show any [Crystools]
- TextEncodeQwenImageEdit
- UNETLoader
- VAEDecode
- VAELoader
- WWAA_AdvancedTextFileReader
- WWAA_BuildString
- WWAA_SearchReplaceText
Download Workflow
WWAA-Qwen-LoRA-Dataset-Creation-Workflow.zip
1302 downloads
