Lora for another OC I came up with.<img src="https://image.civitai.com/xG1nkqKTMzGDvpLrqFT7WA/2dcd7fc4-4adb-4418-9516-43940e2475c6/width=525/2dcd7fc4-4adb-4418-9516-43940e2475c6.jpeg" />My OCs collection: <a target="_blank" rel="ugc" href="https://civarchive.com/collections/201272">here</a>.<h3 id="heading-5037">Lira Hornsby is a cheerful and friendly <s>cow print</s> cow girl who loves to help others with her alchemy and healing skills. Her parents were an unlikely pair - an ethereal high-elf and a minotaur warrior. Lira inherited her mother’s blue eyes and some of her blonde hair. Lira also got her father’s bovine features such as the horns, tail and his black <s>fur</s> hair.</h3><h2 id="heading-2408">Useful information:</h2><h3 id="heading-2409">Pay attention to the wording in your prompts. The result of the generated image should respond to the wording, for example: thick thighs and huge breasts or just thighs and breasts. Or the realistic tag - it can change the face hence give you a variation.</h3><h2 id="heading-5039">Further information:</h2><h2 id="heading-5040">Several issues with this model:</h2>1) COW PRINT, COW PRINT EVERYWHERE! (For some reason according to Stable Diffusion cow=cow print, also let's slap it everywhere 🙃)2) Lira is an incredibly good listener, so much so that she occasionally grows several pairs of ears3) Probably more, so I leave this open, so I can add later...4) Mondays suck5) Sundays are even worse6) ...<h2 id="heading-5040">Results, as with any model, may vary.</h2><h3 id="heading-2410">I have tested the LoRA, but I obviously cannot produce all possible environments, so this model might work perfectly for your personal settings, negative embeddings and other resources used for image generation or it might be terrible. （づ￣3￣）づ╭❤️～</h3>

PulenKompot-Lira Hornsby

First version - it's a variationIt's a LoRA and was trained on NAI diffusion.This version adapts better to different anime models, but deviates (in my opinion) substantially from the dataset - it's most obvious in the faces and body type of the character.Still, I upload it, because this adaptability could be useful to people who use a lot of LoRAs together and like the experiment with different checkpoints.

Files with SHA256: 2d6b2725d86c0c5c2fe75e4bcb260282cef0a24095879e3fe241e6d45a98df4f