Heya! I may take up your request to do a Pony version in some time. I mostly make these LoRAs for myself, so I'm not very keen on taking requests.

However, remember you can always train it yourself as the training data will always be freely available on all of my LoRAs. The training settings I used can be found through the user JustTNP on his LoRA training article (the ILXL GLoRA settings, in particular).

If you decide to train a Pony version of this LoRA, then the only thing I ask of you is that you share the training data as well, just as I have. My philosophy in LoRA training is resources should be made as freely available as possible, that way, whenever the new best checkpoint model comes out and everyone ditches Illustrious, all they have to is copy the training data and train it for the next best checkpoint. I make my LoRAs very slow and steady to ensure I don't make any mistakes and have to go back to fix something. You don't have to say that this training data is mine, but never claim it as your own.

Thanks for understanding!