User LoRA
XL
User LoRA
illustriousXL/NoobAI StabilizerIllustrativeXL and NoobAI are very creative models. So they often output strange/unwanted character poses and styles.I trained this LoRA in the hope of making them more stable and natural through learning more high-quality character-focus images, instead of only relying on tons of negative prompts. And it did achieve my goal.It also can improve the overall output quality. Because in theory it learned more excellent quality images.For demonstration, all cover images have no negative prompts, no artist or style tags.Recommended strength: 0.4~0.8.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 illustrations.Versions/UpdatesNoobAI:ep11 v0.58: More training images (+400). Keep the training parameters as close as to NoobAI base model. Fixed CFG shift issue in previous versions.ep11 v0.30: More training images (+200). Also, this version has stronger effect, recommended strength is 0.5.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:v1.3: Improved dataset tags. Improved LoRA structure and weight distribution. Shoud be more stable and have less impact on image composition. Smaller tensor size (100Mb -> 30Mb).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.