Donations appreciated if you like my work! https://ko-fi.com/doctorshotgun
Alisha Diphda from Tales of Zestiria, trained as a LoRA on Animagine 3.0 Base, intended for inference on Animagine 3.0. Detailed training data is provided in the safetensors metadata included with the lora.
Recommended tags are:
Base Outfit: adarmor, blonde hair, side ponytail, green eyes, armor, neck ribbon, black shorts, garter straps, gauntlets, gloves, greaves
Formalwear: adformal, blonde hair, side ponytail, green eyes, dress, blue jacket, white skirt, thighhighs, gloves, hat, hat feather
Swimwear: adswim, blonde hair, long hair, twintails, green eyes, pink bikini, front-tie bikini top, necklace, sunglasses, armlet, bracelet, eyewear on head
Misc Outfits: dsalisha, blonde hair, side ponytail, green eyes
License:
https://freedevproject.org/faipl-1.0-sd/
Check out my HuggingFace repository:
Description
FAQ
Comments (2)
Wow, these LORAs are a huge step up from your older ones (~8 months ago). Is the improvement from using SDXL? If not, I would really appreciate if you could also update the SD1.5 versions for those of us that are stuck on older hardware. Thank you!
These are trained at 1024 resolution instead of 512 or 768 like the 1.x loras, but XL also does seem to play a role in it. I redid the SD 1.x ones at 1024 resolution as well, but they still don't quite capture the outfit detail as well as the XL loras.
You can check those out here:
Doctor-Shotgun/SD1.5-NAI-Character-Lora-1024 · Hugging Face
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.