Duskfall AI

If you want to donate towards costs and don't want to subscribe:

https://ko-fi.com/DUSKFALLcrew

If you want to monthly support the EARTH & DUSK media projects and not just AI:

https://www.patreon.com/earthndusk

Discord: https://discord.gg/Da7s8d3KJ7

This is our larger training model, on a smaller (203 512x512 images) data set of our own art with a focus on the fact that our art, and any stable/midjourney outputs we included in this are related to our Dissociative Identity Disorder.

Testing space on huggingface: https://huggingface.co/spaces/Duskfallcrew/duskfallai_webui

If you like what you see on our trained (not merged) models go here: https://www.patreon.com/earthndusk

This one SEEMS to be a little more photographic, and there are still a huge lean towards photorealistic catboys, as we test the model i'll get more examples running.

Trained using the MultiModel Dreambooth App, sitting on a friday afternoon doing absolute squat. So yes, it is a MIX/merge of sorts of Stable Diffusion (whatever models were included with that app). We hope you enjoy!

"USE WITHOUT CREDIT" is only if you're merging/putting the model in say a mixed model. Feel free not to credit when you're making or generating images. DO SHARE!! <3 We wanna see what you make!

Keywords are lisdusk1 and lisdusk2

Example below is
"Photographic Portrait , in the style of Larry Elmore, Dungeons and Dragons, Portait, Dynamic Lighting, Colorful , hyperdetailed, Poster Art,Portrait, Futuristic, Dynamic Lighting, Volumetric Lighting, Ray Tracing, Alternative Universe, Retro Futurism, Marvel Comics, X-men, Avengers, Disney, Square Enix, Kingdom Hearts, FFXIV, Flowers, Roses, Utena, Ouran High School, Ranma 1/2, Hayao Miyazaki, Detailed Surrealistic Beauty by Artgerm, Incredible Digital Artwork, Unreal Engine"

And then the bottom one is the same prompt but Protogen 2.2 Anime IMG2IMG

Please note, that the actual prompt does NOT produce copyrighted artwork - LOL, it's just a really cool fusion technique we used to do on midjourney.