This is my second Lora made and my first published one: this adorable male cat from Shangri-La Frontier, who loves an awesome brown bunny.
Should work in any Illustrious based checkpoint. If at strength 1 you feel he doesn't look so accurate, you can increase 0.1 by 0.1, and, something around 1.5 strength it should make him more accurate, also without overfitting.
Trigger word is Aramis, and there are other tags to specify well his clothing and general anatomy, like semi-anthro, digitigrade, etc. (refer to the post's metadata). Also, if you want him not wearing any clothes, you can put nude on positive prompt and/or clothing on negative prompt.
I intend in updating and improving this model in the future, like, making a bigger training with 10 epochs and around 1150 steps, for example.
Description
Trained with:
Checkpoint: AnyIllustrious-XL(for LoRA training) 2.0.0: https://civitai.com/models/823108/anyillustrious-xlfor-lora-training
Program: Kohya SS: https://github.com/bmaltais/kohya_ss
Mixed Precision: BF16
Scheduler: Cosine
Learning Rate: 2e-06
Max Bucket Reso: 2048
Min Bucket Reso: 256
Number of Repeats: 15
Number of Images: 68
Cache Latents: True
Gradient Checkpointing: True
Batch Size Per Device: 5
FAQ
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.