Loading
Du hast einen ComfyUI-Workflow? Erstelle direkt im Browser eine verständliche Workflow-Dokumentation.Workflow verständlich machen.Jetzt ausprobieren ↗
Decided to make a hana uzaki TI which was trained on 400 images from the anime at 10k steps.
Modellquelle
Quellenbeschreibung
Decided to make a hana uzaki TI which was trained on 400 images from the anime at 10k steps.
Works best when you put masterpiece, best quality and even ultra quality in the positive prompt.
Images were made using anything-v4.5-pruned
prompt to generate character: hana_uzaki
Copy the file to stable-diffusion-webui/embeddings
Quellen
1 QuelleModellartefakte
1 ArtefaktQuellenauszüge
2 AuszügeSource context: 2311 downloads · Type TextualInversion · Base model Other
Source context: 2293 downloads · Type TextualInversion · Base model Other
