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

Requested by: @Kryata

Checkpoint trained on and used in examples:

model.ckpt · a1079602570/animefull-final-pruned at main (huggingface.co)

If you want optimal results and don't want to experiment combining styles, use this base model!

It is old but gold, and allows for model training that gets a lot closer to the original, compared to tons of other often used models like AnyLoRA. Otherwise I recommend using simple anime style checkpoints.

I am once again open for suggestions for other styles, SFW and NSFW.

You can always drop model suggestions in the comments, both for styles that are missing on civit or those that are outdated. No guarantees but if it interests me I'll give it a go.

About the artist:

Twitter: ラプトル7☾例大祭お30a (@raputoru7) / X (twitter.com)

Blog: 武器櫃 † ラプトル7 の 聖結界 (fc2.com)

Raptor7 makes almost exclusively illustrations of different Touhou characters, generally depicted as quite slender with lots of detailing in the outfits.

Weights & Prompts

The activation prompt is:

raptor7

Due to the large amount of Touhou characters in the dataset it should be possible to recreate many of them, including the ones in the first 6 example images. Check the example prompts for reference.

Use a weight of 0.8-1.2, adjust to taste and depending on your checkpoint.

The model is optimized for 1.0 while other checkpoints generally work better with a lower weight.

As usual I have included an XYZ plot for reference.

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! 👇