
A drawing style with light oil elements.
How to use
Just prompt away, things this LoRA is good at is horns, pointy ears, demon girls, goat girls, discount draeneis, muscular female, dark-skinned female, various colors of skin, fantasy themes and so on. The "oil inspired" images used for training had a lot of futanari content in it too for those who want to prompt that, see the separate post for lewds. For further prompt tips please refer to the sample images and the uploaded zip of training tags.
Best results for straight txt2img generation on AOM2 seems to be using Euler A or DPM++ 2M Karras with resolutions at 640*768/768*640. If you just use txt2img to hunt for good seeds to img2img you can naturally use lower. Switching between the two samplers can lessen on strengthen the oil-like touch.
MISC
Trained on NAI with a dataset of 120 images all manually cleaned (phew) at 512 resolution.
All sample images are raw txt2img upscales, no inpainting or helper embeddings used.
The training was done on a small subset of the artists images, as such the output of this LoRA is not an accurate representation neither in quality or style of the source material. Go support the artist Deathmask if you want the good stuff.
描述:
Initial release
训练词语:
名称: Deathmask Training Tags.zip
大小 (KB): 54
类型: Training Data
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
名称: Deathmask-NAI-LyCORIS.safetensors
大小 (KB): 29679
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success