When using the Civitai Generator, if you do not enable Enhanced Compatibility, abnormal results will be generated!
Setting the weights to approximately 1.0 is recommended.
Activation tag : minecraftskin
Recommend Prompt
Positive: minecraftskin, 1girl, solo, simple background, green background, full body, no mouth, no lineart, standing, (tag), (masterpiece, best quality, absurdres, highres, very awa, newest:1.2), Negative: worst quality, worst aesthetic, bad quality, artistic error, old, ealry, mammal, anthro, furry, feral, semi-anthro, lowres, (bad), off-topic, multiple views, comic, very displeasing, displeasing, chromatic aberration, (extra digits), fewer digits, fewer, text, error, missing, jpeg artifacts, artist name, signature, twitter username, username, watermark, scan, unfinished, [abstract], variations, (wrong hand), no pupils, Old Guide
Always use the NoobAI 1.1 Checkpoint! (The test results show that the character LoRA works well!)
You must download the Training Data and use the workflows and images included in the zip file!
It is believed to be capable of generating only female characters, as it was trained exclusively on them.
It is possible to create it with the Civitai generator, but it won't be as good as using ComfyUI or other local generators.
Description
This version removes the previously problematic datasets and fixes issues found in the remaining data. It has also been transitioned to a Vpred-based system. Additionally, a new training configuration has been applied, and unlike the previous version, you can now create a Skin simply by entering a prompt!
FAQ
Comments (4)
it compibialited with other loras characters or not?
its looks so great as well
As far as I know, there’s no issue at all.
Are there any chances that you will make a 128x128 skin generator? If not, then if it's not a secret, how did you set the tags? Manually or somehow automatically? How many images did you need in the dataset?
I've put together a dataset of about 1,000 skin resource images. You should be able to train it if you tag the 3d renders of those skin resources using the wd14 tagger.





