
My first stab at a style LyCO, in this case one of FlyTrapXX's style. I found I got the best results when keeping the weight around 0.9-1.0, it just depends on the LoRA being paired with it. You also may need to adjust other LoRA weight quite drastically depending upon the style burn. I also included (flat color, simple background):1.5 in most of my images as it helped the style become more recognizable.
Please, give me any tips you have or just comment down below! I need a bit of help adjusting things and making styles better. v2 will definitely have a manually tagged dataset, but the amount of images make it much tougher to tag everything.
Enjoy!
描述:
Trained on NAI with an automatically tagged dataset. Best when used at a weight of 1.0, but adjust depending upon your needs.
训练词语: flytrapxx
名称: flytrapxx_style_v2-31.safetensors
大小 (KB): 92780
类型: Model
Pickle 扫描结果: Success
Pickle 扫描信息: No Pickle imports
病毒扫描结果: Success