- Oct 26, 2017
- 6,216
- 18,970
Hi! Are these images generated from single checkpoints only? Did you use enhancement afterwards? Or is it enhanced with one or two Lora?Experimenting with a new style - I like the vibe, but it struggles with dynamic posing and perspective prompting, and the faces are very generic. It's also seriously resource intensive by PonyXL standards
Dumping a bunch of pics cos I'd appreciate feedback. I think the only way to really solve the promp issues is to build a checkpoint. That's a lot of works so I don't want to bother unless folks like it as much as I do ^^
View attachment 4559424 View attachment 4559425 View attachment 4559426 View attachment 4559427 View attachment 4559428 View attachment 4559429 View attachment 4559430 View attachment 4559431 View attachment 4559432 View attachment 4559433 View attachment 4559434 View attachment 4559435 View attachment 4559436 View attachment 4559437 View attachment 4559438 View attachment 4559439 View attachment 4559440 View attachment 4559441
Hi! Are these images generated from single checkpoints only? Did you use enhancement afterwards? Or is it enhanced with one or two Lora?
Also, if it's not a secret, is it generated locally or online?
Thx