This is a LORA trained on the style of Kan Liu. There's no trigger tags or anything, so just prompt normally and have fun. The XL version usually does a good job staying out of the way on composition details and sticking to just handling style stuff. Unlike the V1 and v2 models, you can even push the faces out of the typical Kan Liu range and into the wild PonyXL range by simply prompting for the expressions you're familiar with from the model. This might be a negative for some, but I personally enjoy the flexibility I got out of v3.
The 1.x versions are a bit different, mostly down to stiffer composition and worse faces and hands. You can still have fun with it on SD1.x anime models, and v2 does some wonderful stuff when prompted for ink, shadows, hydrokineses, and chromatic aberration, but I can't really say as many nice things about it as the XL version.
Anyway, check the version descriptions. Minutia about each version is included there.
Description
Reduced pose-locking and unprompted water issues. Distant faces are also improved over the last release. Should now be more responsive to tags like ink and hydrokinesis. and just generally do a better job of getting out of your way while still adding to the final result when this isn't the only style element you want to use. In testing I found the best range for this version to be between 0.7 and 0.9. Still strongly influences outfits and hair color/style, which should hopefully be fixed by the next release.
FAQ
Comments (5)
Which model with hash HDDDD did you use for your great renders ?
It's a personal merge, but a major part of it's Hentai Diffusion.
I don't feel comfortable releasing that one, since one of the models in there has a pretty gnarly license, so I'll just list the models contained in it. Hopefully people can forgive that I don't remember the ratios for everything; I put it all together months ago.
Hentai Diffusion H17
Dreamlike Diffusion v1
JWST Deep-Space Diffusion
Double-Exposure Diffusion
Thanks, we understand. Where can we find the Hentai Diffusion ? Here in CAI doesn't exist (so says the search engine).
@ritcher1 It's over on their Github page currently (seen it move from HF to Github to a dedicated site and back, so this link might not stay correct): https://github.com/Delcos/Hentai-Diffusion
I find it's a little scuffed out of the box, but merging it with Dreamlike did a lot to impart a nice base style on it and restored a lot of artist tag functionality. The rest of the stuff in the merge is specifically for adding good sparkles and a double-exposure effect when I want it.
@LeTree Perfect, thanks !