>>70633444Ok I'm finished. Same deal as previously
Grid. No upscale/inpaint.
Some attached pictures with upscaling
Images:
https://files.catbox.moe/c809pk.pnghttps://files.catbox.moe/34ibty.pnghttps://files.catbox.moe/yqf5qs.pnghttps://files.catbox.moe/qha3kw.pnghttps://files.catbox.moe/aevc2o.pnghttps://files.catbox.moe/gymkbj.pnghttps://files.catbox.moe/t9ku5x.pnghttps://files.catbox.moe/z0w422.pngLora:
https://files.catbox.moe/pvf1kk.safetensorsSide Note:
Her face came out "stiffer" compared to my other loras. You'll need to prompt facial expressions or it'll mostly be the expressionless/dead face type of look. The data set that I used from that other anon is a contributing factor, but I'm not specialized in training indie chuubas with small datasets anyway. Did the best I could!
