
Patreon
Get early access to build and test build, be able to try all epochs and test them by yourself on Patreon or contact me for support on Discord
Recommended settings
-
Make sure you are aware on the usage instructions of LORA
-
Sampler: DPM++ SDE Karras (Recommended for best quality, you may try other samplers)
-
CFG Scale : 5 to 10
-
LORA weight for txt2img: reccomended 0.2-0.5.
Insutrctions
My advice is to start with posted images prompts then change them, in any case you will find different keywords for each model (wich should be used with a weight of 1.2-1.3).
Also in version description on the right you will find more details about each hairstyle model.
Enjoy!
If you'd like to support my efforts and have access to early and bleeding edge builds, please think about joining my Patreon: https://www.patreon.com/betterCuteAsians
描述:
As you can see many of my images have a perfect hime cut hairstyle not showing the fronthead, in those images i used my model in inpaint, first i generated the images as i wanted, then i sent to inpaint, masked the fronthead, checked fill checkbox and wrote in the prompt only this: "(long_hime_cut_hairstyle:1.3),<lora:long_hime_cut_hairstyle:0.5>" and then generated to have a perfect hime cut hairstyle.
I didn't do that for all the images, some of them are just generated, but they are not perfect hime cuts. So, remeber to inpaint if the generations is not how you want it.
Remember if you are inpainting an image after the other that before sending to inpaint your image you have to close the last image from img2img or in the new generation it will remember the old masked part other then the masked part of your new image, doing a mess.
Also many of the images are generated using controlnet poses packs that you can find in my page.
Works also with anime.
训练词语: long_hime_cut_hairstyle
名称: long_hime_cut_hairstyle.safetensors
大小 (KB): 147571
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success