
This workflow uses qwen3 vl to create a 10 prompt list each time you run it and qwen edit 2509 to generate the images. If you want the wan 2.2 refinement, leave the group on. This workflow is perfect for generating the perfect dataset to train a lora.
angela rose
2025-11-20 Update
Follow
angela rose
2025-11-20 Update
Follow
Workflow introduction
This workflow uses qwen3 vl to create a 10 prompt list each time you run it and qwen edit 2509 to generate the images. If you want the wan 2.2 refinement, leave the group on. This workflow is perfect for generating the perfect dataset to train a lora.
Nodes Information
25
CLIPLoader
CLIPTextEncode
KSampler
LoadImage
LoraLoaderModelOnly
SaveImage
UNETLoader
VAEDecode
VAEEncode
VAELoader
AILab_QwenVL
BetterFilmGrain
CFGNorm
CR Prompt List
Fast Groups Bypasser (rgthree)
ImageScaleToMegapixels
ImageScaleToTotalPixels
ImageSharpen
LayerUtility: PurgeVRAM
MarkdownNote
ModelSamplingAuraFlow
ModelSamplingSD3
TextEncodeQwenImageEditPlus
UpscaleModelLoader
easy showAnything