Follow me on X/Twitter for exclusive content: https://x.com/Citron_Legacy 😁
Please checkout all my Pokemon Loras: https://civarchive.com/user/CitronLegacy/models?tag=pokemon
Description
Update on 05/27/2024 - PonyXL released!
Trained on the same dataset as v4
FAQ
Comments (11)
Wow ❤️
Lol Zoe kind of counts as a "off-brand" Pokegirl?
She does! 😍
She's pretty 🙂
Thanks! I'm pleased with how the PonyXL version turned out.
@CitronLegacy You're welcome. Your models are so good :D
Poor gal, just wanted out of the crime business and had everything ready to dissappear along with her BFF grizzbolt then a caveman waving a triple bow mounted on a pig comes and kills her. Really hope we can spare her later, thanks for the model my dude
I've only read some of her backstory but she seems like a good person. She will live on in fanart/AI-Art!
Glad you like the model!
I seem to be missing something because my images are coming out as blobs of colors with maybe a slight resemblance to a character. how are all those loras used without them being called in the prompt? even adding them in my images are still coming out horrible.
Hi.
The images I posted were all generated on CivitAI so the normal Lora invocation that you see in Automatic1111 (or whatever software) is not present in the prompts.
1. You can use the CivitAI generator to make images: I go out of my way to ensure all my images on this model are easily reproducible
2. You can learn how to translate CivitAI generation prompting into Automatic1111 but this may take some time. Theoretically this is as simple as adding <lora:ZoeRaynePDXL:0.7>, to your prompt (but sometimes PonyXL can read the Lora name as part of the prompt and cause issues). Try invoking this Lora like you would normally and just know that CivitAI/ComfyUI prompting looks different than Automatic1111/Similar.
@CitronLegacy ahh ok, that explains why some loras don't have the calls for them, I always worked in automatic 1111. I did try adding the <lora:ZoeRaynePDXL:0.7> stuff to the image and it at least sort of started drawing the character, but halfway the image would suddenly get horribly blown out and bright with really low detail. I'm starting to wonder if there's some kind of tag causing the whole image to blow up. I tried the prompts with just the prompt and no lora and it still caused the image to be a bright blob of colors. I seem to sometimes run into this when I alter some prompts from loras for XL models and haven't figured out why. I'll keep messing around and see if I can get something to work, Thanks for the info.
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.