Type | |
Stats | 209 |
Reviews | (3) |
Published | Jul 6, 2023 |
Base Model | |
Usage Tips | Clip Skip: 2 |
Hash | AutoV2 452F9E1E59 |
Version 3 - This is the final version which adds an additional model to version 2 in order to solve the problem of not being able to create male subjects. I mixed in the BB Male Toon Mix model by ewolf to get the male characters. It does create male characters now, but sometimes they come out really effeminate. Still, that's not a bad thing depending on the image output you are looking for. The merge did remove some of the Anime Lineart Style Lora design from the final output, but I still think the design is close enough to be a good successor to Version 2.
Version 2 - This is an stylized model that mixes version 1 with my Anime Lineart Style Lora at 0.4 strength. It provides a different art style, but the model isn't able to create male subjects at all. I'm currently working on a version 3 which should allow for male subjects, but the style will end up being different since I'm merging a model designed specifically for male subjects into version 2.
This is a merged model of three Alpha Lyrae models: LyricED, AvaloriaED, and Terrane by EroticVibes. I loved the styles for all of these models and was curious what would happen if I merged them together. I like the images this model creates due to their unique style and mix of sexy and cartoony.
I've tested for NSFW and it works, though the images are a little too cartoony, so it almost looks like Rule 34 porn using western cartoons.
This model doesn't need a VAE. I've baked in Blessed2 VAE.
Note: version 1 and 2 don't work so well with male characters. I posted some sample images in the gallery for when I attempted to make fantasy images with the prompts for a muscular man and many of the images came out as effeminate men or simply as women. Haven't tried it with character loras, so it may be more successful at making male subjects using loras.
All of the images I upload for this model are unedited and don't use any special extensions like ControlNet. I do not do any inpainting or img2img to correct errors. Why? Because you deserve to see how the model works without any touchup or special extensions like ControlNet. You can surely get better images by putting in the extra work, which I do for my personal images. The samples are just meant to showcase what the model can do raw with nothing but prompt engineering.