Type | |
Stats | 68 324 |
Reviews | (10) |
Published | Apr 14, 2024 |
Base Model | |
Training | Steps: 998 Epochs: 10 |
Usage Tips | Clip Skip: 1 |
Hash | AutoV2 CCB6E7E580 |
This LoRA is intended to help generate better American flags πΊπΈ.
Version 2.0 (SDXL):
(Updates: In this version I slightly modified the training words and trained the model on some of the finer details of the flag.) More testing is needed, but it looks like this version may require use of a higher weight in prompts. example: ((US Flag:1.2)).
Words to prompt:
us flag, american flag, 13 stripes, 50 stars, stars and stripes
In negative prompt:
bad flag, too many stripes
___________________________________________________________
Version 1.0 (SDXL):
Words to prompt:
us flag, stars and stripes, american flag
In negative prompt:
bad flag, too many stars, too many stripes, bad render
All feedback, ideas, and shared generations are greatly appreciated and will contribute to future creations!