Don't use it,it‘s too old and not enough train.

recommend image scale less than 1.3/1 ,such as h1280*w1024,otherwise multiple heads/waists/necks may appear in the image if over image scale.

推荐图片比例小于1.3/1

2.0 shouldn't require trigger word , and even if they are used, there shouldn't be any harm.

2.0版本应该不用触发词了，要用也没什么坏处。

recommended more and more prompts,sdxl model requires them.

建议prompts尽量多点，sdxl底模需要

Train base the shikianimexl and Stable-Diffusion-XL-Anime by bdsqlsz after MBW.

Trigger Word: nebulaewisdomanimestyle

触发词 nebulaewisdomanimestyle

recommend prompts:

breathtaking,nebulaewisdomanimestyle,your prompts. gorgeous,key visual, vibrant, studio anime,award-winning, professional, highly detailed,high budget, cinemascope

negative prompts:

noisy, blurry, grainy,text, graphite, abstract, glitch, deformed, mutated, ugly, disfigured, (realistic, lip, nose, tooth, rouge, lipstick, eyeshadow:1.0)

This is a beta model, so best to use Trigger Word ,beacuse I added the test style lora with trigger word in the last step of the merge.

测试模型，最好用触发词，因为合并的一些画风lora里我塞了触发词

I don’t want to train it again.It Just Works!

懒得重新炼了，又不是不能用!