If you enjoyed this model, why don't you check out my other works? I put a lot of effort into correctly recreating artist styles and making the occasional character LoRA.😊

This model was trained using the collab/guide by holostrawberry [Guide] Make your own Loras, easy and free - colabs | Stable Diffusion Other | Civitai

I noticed that we have multiple Shuten Douji LoRA but Ibaraki doesn't get any love. Furthermore I wanted to see how well you can include multiple versions without creating specific prompts and I've gotta say, I prefer it this way.

Because fanart is limited for some of her versions, I recommend experimenting with the later prompts, especially with prompts changing the hair color.

Weights & Prompts

This version is pretty strong overall. I recommend a weight of 0.7-1.0.

At around 0.7 the clothes become less accurate, below 0.6 the hands are less likely to be red and below 0.4 you lose most other details (check out the XYZ plot for details).

The only necessary prompt is:

ibaraki_douji  

Basic version (add details like "fire" or "rope" to get closer to the ascensions):

 yellow kimono,cleavage, very long hair, japanese clothes, wide sleeves, off shoulder, floral print,  beads, sash

Swimsuit:

very long hair, bikini, japanese clothes,kimono, off shoulder, black bikini, sandals, front-tie top, front-tie bikini top, blue kimono

Swimsuirt 2nd ascension (one-piece swimsuit):

twintails, red hair, hair flower, white one-piece swimsuit, japanese clothes,pink kimono, off shoulder

Swimsuirt 3rd ascension:

gloves, bare shoulders, twintails, very long hair, pink hair, japanese clothes, shorts, black gloves, elbow gloves, midriff, white crop top, gradient hair

The model is still flexible so clothe her however you want, as a jiangshi, maid or casual.

The only thing that is really difficult to do is change her horns or facial features without going real low with the weight. Check out the examples for inspiration.

This was certainly a fun but exhausting one.

Please leave a review if you enjoyed this model or have problems with it. This helps me with improving future models and finetuning them in general. Give it a try and throw some images down below! 👇