A SDXL model for generating image
Description
FAQ
Comments (7)
this is so good, I'd love to see a pony merge of this. I can only dream though, you do you.
Give a link to a specific or similar model. I'll see what i can do. I don’t promise that I’ll do it, but if I like the idea, maybe I’ll do a merge.
@devpilgrim if you go to the models page and set your model filter to pony, you will see a whole bunch of pony merges and how they turned out. the nice thing about the pony model is that it seems to have a higher token capacity, and can follow prompts a little better. pony is highly pornographic in a lot of cases because of that, however, some people are using that token space for other ideas.
most merges just use the pony diffusion 6 model as the base model, and train onto it.
@goronplus Thanks, that's about what I thought. As for pornography, I definitely won’t do that, especially for ponies :) But it’s a worthwhile idea to see what token expansion blocks are in these models. I think I'll start disassembling pony diffusion 6 this weekend...
@devpilgrim Earth Animix PonyXL doesn't seem to have a focus on Pornography, you might try that.
@goronplus Trial version Pony ready: https://civitai.com/models/348373/pilgrimponyxl You can try. I hope for comments :) It will be interesting to know your opinion
@devpilgrim I'm trying it, and it seems to be able to add nice details that I wouldn't otherwise be able to, because other models seem to ignore certain requests. loving it so far, keep training it with new stuff!









