DoRA goal of which is to deliberately degrade fidelity of your generation to add rough look. Or give you a nice, thicc lineart.
Instruction:
This is triggerless training, meaning lineart/sketch is base of weights and you don't need to prompt them, but you can if you really want to.
Negative greyscale, monochrome, if you want it to be colored.
ko-fi - https://ko-fi.com/anzhc
描述:
训练词语:
名称: sketchDiffusionDora_v10.safetensors
大小 (KB): 501163
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success
