I am not sure the reason; I guess it's because they pruned some data from SDXL 0.9 and then released SDXL 1.0, I used the same dataset, same network Rank, same network alpha to train the same character Lora with SDXL 0.9 and SDXL 1.0, and the result is that the SDXL 1.0 lora generated more bad anatomy images. Here is another file:
https://civarchive.com/models/117366/character-ganyu-genshin-impact-sdxl09
(the sample images are picked images with less problems)
Refiner is not necessary.
Description
FAQ
Comments (6)
Do you have tutorial for training lora on 1.0?
not recommend use 1.0 to train.
I appreciate your effort, but these look whacked. Wait for the Anime fine tuned model to come up, it's due.
try the 0.9 version
https://civitai.com/models/117366/character-ganyu-genshin-impact-sdxl09
Details
Downloads
652
Platform
CivitAI
Platform Status
Available
Created
7/27/2023
Updated
5/13/2026
Deleted
-
Trigger Words:
genshin-ganyu
bare shoulders
blue hair
purple eyes
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.