NoobAI/illustriousXL Stabilizer

NoobAI/illustriousXL Stabilizer-龙跃AI
NoobAI/illustriousXL Stabilizer
此内容为付费资源,请付费后查看
10
付费资源

illustriousXL/NoobAI Stabilizer

I honestly don\’t know what to name this LoRA.

IllustrativeXL and NoobAI are very creative models. So they often output strange/unwanted character poses and styles.

I trained this LoRA because I want to make the character pose and style more stable and natural, through learning more high-quality character-focus images, instead of only relying on tweaking tons of negative prompts. And it did achieve my goal.

So far the training dataset contains ~1k hand-picked images that have:

  • Excellent quality. (≈ Wallpaper level).

  • Character-focus. (The characters take up most of the image.)

  • Clear characters. (Clear character lines, no obstruction, no blur effect, etc.)

  • Natural poses. Natural body proportions. (No exaggerated art, chibi, jojo pose, etc.)

  • 100% 2D color illustrations.

Therefore, this LoRA will tend to generate images with the attributes above.

Recommended strength: 0.4~0.8.

This LoRA is not overtrained. Which should has good compatibility and can preserve base model\’s original creativity and style.

For demonstration, all cover images are generated using only the base model, no negative prompts, no artist or style tags, etc.


Versions/Updates

NoobAI:

  • ep11 v0.58: More training images. Change the training parameters as close as to NoobAI base model. Fixed CFG shift issue in previous versions. Very compatible with NoobAI v-pred verison, even in dark env, tested on vPerd09R, reached my 9/10 expectation.

  • ep11 v0.30: More training images.

  • ep11 v0.11: Trained on NoobAI epsilon pred v1.1. Using new stuffs same as illus v1.3.

  • ep10 v0.10: Trained on NoobAI epsilon pred v1.0.

illustriousXL:

Note: illustriousXL version will not be updated. More and more checkpoints are based on NoobAI or merged both illustriousXL and NoobAI. It\’s better to use the NoobAI version above.

  • v1.3: Improved dataset tags. Improved LoRA structure and weight distribution. Shoud be more stable and have less impact on image composition.

  • v1.0: Init version. Trained on illustriousXL v0.1.


This Lora is only published on C.i.v.i.t.a.i. If you see this sentence on other websites, it means it is plagiarism.



© 版权声明
THE END
喜欢就支持一下吧
点赞19 分享
评论 抢沙发
头像
欢迎您留下宝贵的见解!
提交
头像

昵称

取消
昵称表情代码图片快捷回复

    暂无评论内容