V3.0 has been released.

Some of the training files have been replaced and the shading of the nose has been adjusted.

Also, a friend of mine discovered a prompt that can be generated without a hook.

If you enter [nose nostril] instead of the trigger word, the probability of it being generated increases significantly.

Due to certain circumstances, I updated to 2.1. I made some small improvements.

V2 released.

I changed the image and did some training.

The basics are the same as V1.

Please try to enter the prompts carefully as it seems that it is easy to get ugly images.

The PONY version has been released. The training content is the same as SDXL Nose hook Anime style V5.

I've only been using PONY for a short time, so I haven't been able to find a better prompt yet.

Depending on the prompt, my drawing style can change quite a bit.

The base model used for training was the Pony Diffusion V6 XL.

Not suitable for real picture.