If you enjoy my content, consider supporting me on Ko-Fi! ☕
Requests are always welcome in my channel in the Furry Diffusion discord! 📝
--v1 XL-- [PXL, LoCon]
Trained & tested on PonyXL v6.
Spooki hellhound gal for spooki month - made on request! (Because we definitely needed yet another Loona model)
Recommended positives: hblna, white eyes, red sclera, slit pupils
Outfit positives: penttop, hotpants, spiked collar
Accessory/Extra positives: ear ring, fingerless gloves, pentagram harness
The clothing may also be controlled with "torn topwear" and "torn bottomwear" if you want damaged-looking attire, or if you don't want it "torn clothing" in the negatives will help.Description
FAQ
Comments (6)
Seriously? Another loon lora that's been on civitai for a while now???
well, it's not up to you to blame people for doing something that they want to do...
BUT I have to say with this LORA - even at 0.7 strength - it decreases the image quality.
Yiffymix_v52XL for example was doing a better job on it's own as with the LORA in creating a good Loona.
@nightdancer495 Hmm, that's odd - It might have to do with how Yiffymix works, it's worked just fine for me on base Pony. And while I'm aware Loona is quite well represented in most datasets, this was simply made for someone who wanted a lora for her made by my own hand, haha. If anything, you could mix-and-match between the base model's interpretation and the loras' for different aspects.
@Valstrix yiffymix is a better detailed model as pony for example and yes Loona is well presented in most furry models, because just have to learn to prompt for stuff as instead using a LORA for everything ;-)
some char-LORAs are useful, if the base model isn't trained on the subject.
i don't mind another loona lora, but i haven't seen a feral loona yet.
Before generating an image, replace the word anthro with feral..
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.