Andrea V1
SDXL 1.0 Standard LORA
DownloadAndrea V1
V1 - Still early days; Face is consistent only with very specific inputs. You can try increasing LoRA strength to 1.1 if there is not enough familiarity.
Just a character that I was interested in out of a random generation. Now trying it out with SDXL as the base Model.
The LoRA works best between 0.6 - 0.8 in strength.
Do let me know your experience with it, as I'm fairly new to training models.
Would love to improve this and make Andrea a well tuned model.
PS: Funnily enough, I've not been able to get this to work on the civitai generator; The generator waits on, and then drops out.