Trigger words:
human loona \(helluva boss\), loona, long hair, grey hair, undercut, white sclera, red eyes, makeup, eyeshadow, black lips, piercing, ear piercing, black choker, bare shoulders
Outfit:
off shoulder, crop top, shirt, torn clothes, navel, midriff, black gloves, fingerless gloves, black skirt, black shorts shorts
Tested and used with Ikastrious (NoobAI-XL):
https://civarchive.com/models/874216?modelVersionId=1229496
v1.0 V-Pred: IMPORTANT: This LoRA is trained as a V-Pred model and is fully compatible with other V-Pred models. For the best results, make sure to use it with a V-Pred checkpoint.
Tested and used with Noobai-Cyberfix/Perpendicular-Cyberfix:
https://civarchive.com/models/1001795/noobaicyberfix-updated-vpred-eps?modelVersionId=1140938
For on-site generations, use my Tensor Art page: https://tensor.art/u/830771749061779884
Order a commission here!
https://ko-fi.com/c/38aa4a973d
Support my work by buying me a ko-fi:
Description
FAQ
Comments (2)
What's the different between V-pred and non-V-pred checkpoints? The outputs seems similar, but people always say you have to be sure not to mix Lora and Checkpoints that don't match.
I do not know how to explain it well, but for me V-pred generated higher quality and more screen-accurate images. For example, when you use the v1.0 Loona model, you would use a non V-pred checkpoint like WAI-NSFW-illustrious-SDXL or Hassaku XL (Illustrious), because I used an Illustrious checkpoint as the base model to train the images. I used a V-pred model as the base model for the updated version and that is why the images are more screen accurate and higher quality. If you're gonna use the v1.0 V-pred version, make sure you use a checkpoint that has V-pred in the name. I've seen people using the wrong checkpoints when using my V-pred LoRAs. I always link the checkpoint i've used when generating the sample images. Hope this helps!
Details
Files
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.