Lokr trained with Hizake's style.
420 newer images were used for training, mainly human females, but males and non-human characters can also be generated. The style is characterized by soft colors and delicate eyes.
One thing to note is that the nsfw images released by the artist are all censored, and I can't remove them, so I train with them. If you don't want to see censored, be sure to add "uncensored" to the positive tags and "censored" and "mosaic censoring" to the negative tags.
This lora can be used without a Trigger Word, but it still has a Trigger Word: hizakestyle. It makes almost no difference whether it is used or not. If you want to use it, use (hizakestyle: 0.6~1.0).
Weight: 0.6 - 1.0. 0.7 and 0.8 are recommended.
Support the Artist: https://lusterizer.fanbox.cc/ https://fantia.jp/fanclubs/94720
If requested, I will delete this lora.
使用Hizake的画风训练的lokr。
使用420张较新的图片进行训练，主要内容为人类女性，但男性和非人类角色也能够生成。该画风特点为柔和的色彩和精致的眼睛。
需要特别注意的一点是：该画师发布的nsfw图像，均带有马赛克，我无法去除它们，所以带着它们进行训练，如果你不想要看见马赛克，则务必在tag中加入uncensored并在负面tag中加入censored和mosaic censoring。
本lora无需Trigger Word即可使用，但仍然带有Trigger Word：hizakestyle，是否使用它是几乎没有区别的。若要使用则用 (hizakestyle: 0.6~1.0)。
Weight: 0.6 - 1.0，推荐为0.7和0.8。
请支持画师: https://lusterizer.fanbox.cc/ https://fantia.jp/fanclubs/94720
如果被要求，我会删除该lora。
Description
This version of lora is based on Noob-V-Pred-1.0.
(All sample images are generated with Noob-V-Pred-1.0 using ComfyUI.) (Note: In the sample images, the weight of the quality-enhanced lora is lower, at 0.35)
该版本lora基于Noob-V-Pred-1.0。
（示例图像全都用Noob-V-Pred-1.0，使用ComfyUI生成。）（注意：示例图片中，提升质量的lora的重量都是较低的，为0.35）