V2 Info:

R-18 preview on CitrineDreamMix is here: https://www.pixiv.net/en/artworks/113044319

Model of Redeemer boss from a game Vindictus (Mabinogi Heros) - Laura.

Model was trained on 80 images, with 5 repeats in 10 epochs. 2000 steps in total.

Model was trained with 2 Checkpoints: MeinaMix and CitrineDreamMix. Main one is MeinaMix.

Activation tag: Laura_(vindictus)

Appearance prompts: short hair, fake animal ears, purple eyes, purple hair, cat ears, facial mark, forehead mark, lips

Outfit prompts: necklace, cleavage, jewelry, armlet, bracer, ornament, choker, angel, angel wings, feathered wings, see-through

Helpfull prompts: 3D, golden ornament, golden choker (warning - it can change many things around to a golden, so use it only if the choker and the mark is purple), large breasts/medium breasts (you can change their size), realistic

For other prompts, check the images on post.

Reccomended weight is 0.7-1. On 1 may appear something from her training images, like a grey floor from gm_flatgrass or collumns from her battle area. Reason is that she was trained on a 3 types of images - screenshots of her ragdoll from Garry's Mod, screenshots from original Vindictus (i was doing her solo and sometimes taking a photos). and images maked on V1. Technically, V2 = V6, 'cause i was trying to train her on other things too. But versions 2-5 is not good enough for me to show you.

Model is trained on 1024x1024 resolution, so it's better to use 512x512, 768x512 and 512x768 resolutions to make a images, then use img2img with denoising 0.5-0.55 to make it x2 bigger.

If you see that her choker, armlets and facial mark is purple and not golden, it's ok. The reason is that it was purple in GM and golden in original. But the chance of appearing a purple ones is much lower.

For unclothing use a tags: NSFW (will change her at all), nude, nipples. She trained for some poses, so you can try many of them.

The goal of training was to make her looks 90% similar to her game model. 10% of her appearance (on the "belt zone") was reduced. If you want to see the big ass and pantiess like a diaper - better train a new model. Yeah, i saying it to you, Jonathan)

V1 Info:

Sample model, which was used to train a V2. Have a big problems with a backgrounds.