At the request of one of the users of my models, I created a Ponyxl model in the style of Pilgrim_2D_SDXL. This is not a merge. One of the Ponyxl models was taken, only the blocks were cut out of it, and instead a data set was added on which Pilgrim_2D_SDXL was trained.
The result is not quite what I expected yet, but in the future I will still work on the model.
If you are interested in this model, do not hesitate to write comments, leave your works and like.
Description
FAQ
Comments (3)
I didn't want to put a thumbs down, so just leaving my feedback - I really love the art style, but this particular model has a lot of the "Pony" functionality stripped - that is, the robustness of Pony to detects most image using texts. I'd be keen to see an update at some stage that addresses that. Either way, thanks for your amazing work. :)
Yeah, my thoughts exactly. Maybe it's easier to train a lora in this style than to finetune or merge pony?
Well, the functionality of the model is really reduced here. I removed some of the tokens that relate to pornography. Also, this greatly influenced the relative weights in the model itself. This is only a trial version, I will continue to understand the Pony structure and refine the model.