Bumblesteak Style版本v1.1 (ID: 57099)

Create characters in the style of Bumblesteak

How to use
No trigger word needed. Just prompt away. The dataset was fairly limited and this LoRA seems to work best with monster or animal girl tags, dark-skin and fantasy style content. It also seems to be pretty good at "black sclera" which often can cause problems prompting otherwise.

Don't forget to review and upload your own images, it's always fun to see what you're creating!
Like my LoRAs and want more? Please consider a small tip towards my electricity bill and further training >.<

Misc
512 resolution trained on NAI without noise offset option for best LoRA compatibility.
All images are untouched raw txt2imgs, no inpainting done, for quality transparency.
Recommended model is NAI or any other with good support for drawing styles.

描述:

Retrained with new parameters. Should be more flexible while at the same time being better at reproducing the style especially on AOM2. Still not perfect, hands and faces in zoomed out shot's will likely need some img2img/inpainting work.

训练词语: