This LoRA is trained for 2 models, heavy-line.safetensors for v1.1 version of NoobAI XL, and heavy-line-mmh.safetensors for Vpred 1.1 version of MiaoMiao Harem.

The dataset used to train this LoRA is scraped using LagPixelLOL/aisp, containing a total of 578 images, a total of 3 artists are used.

Big thanks to the artists for the very cute styles :3.

To use this LoRA, you can go without a trigger word, which will use all 3 artists' style together, or you can choose to specify which artist's style with a trigger word, note this model is mostly a foot model.
pixiv @くろやくそく: hei yksk
pixiv @leonzo030: leonzo
pixiv @フリザ: efreezerarts

This LoRA is trained using kohya-ss/sd-scripts, with rank 32, alpha 16, learning rate 1e-4, for 192 epochs with a total of 5184 steps, using a B200, took approximately 6 hours.

If you have any questions, suggestions, or just want to talk to me, you can add me on Discord with ID @v2ray.