
After a few iteration, I think I'm starting to get the gist of Lora creation, even if it's far from perfect.
This one was trained on about 100 images, extracted for gelbooru, with a bit of manual images & tags pruning.
Prompting should include "megumin", and can be equiped using :
witch hat, cape, staff
The staff itself might get a bit weird, but inpainting it using only "staff" as a prompt, without mentionning megumin, should give better result.
描述:
After a few iteration, I think I'm starting to get the gist of Lora creation, even if it's far from perfect.
This one was trained on about 100 images, extracted for gelbooru, with a bit of manual images & tags pruning.
Prompting should include "megumin", and can be equiped using :
witch hat, cape, staff
The staff itself might get a bit weird, but inpainting it using only "staff" as a prompt, without mentionning megumin, should give better result.
训练词语: megumin,witch hat,cape,staff
名称: SDXL - MeguminXL_v2.5.safetensors
大小 (KB): 223108
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success