Use "fox ears, multiple tails, korean dress" for variety.
Description
LoRA created.
FAQ
Comments (2)
this lora was linked from the kohyass github. was hoping there would be a config to look at.
Hey, thank you for your interest. I think there should be metadata saved which you can read. I don't clearly remember the details of training but it was very vanilla and simple stuff. The key is to (IIRC) use the base sd 1.5 model for LoRA training, not to overfit data and regularize a bit. A LoRA trained on base SD 1.5 will work on almost all SD 1.5 fine-tunes. Hope this helps
Details
Downloads
864
Platform
CivitAI
Platform Status
Available
Created
6/27/2023
Updated
6/11/2026
Deleted
-
Trigger Words:
Ahri
fox ears
fox tail
multiple tails
korean dress
detached sleeves
whisker markings
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.