๐๐๐ฝ๐ฝ๐ผ๐ฟ๐ ๐บ๐ฒ https://www.patreon.com/scarxzys
๐ฟ๐ฒ๐ฐ๐ผ๐บ๐บ๐ฒ๐ป๐ฑ๐ฒ๐ฑ ๐๐ฒ๐ถ๐ด๐ต๐: 0.7~1.0
๐๐ฝ๐ฝ๐ฒ๐ฎ๐ฟ๐ฎ๐ป๐ฐ๐ฒ:
olivia vel vine, ahoge, very long hair, black hair, hair between eyes, purple eyes, animal ears, cat ears, cat girl, animal ear fluff, cat tail, tail ๐ข๐๐๐ณ๐ถ๐:
cleavage, cleavage cutout, navel, navel cutout, bodysuit, brown bodysuit, clothing cutout, jacket, collared jacket, white jacket, belt, black belt, boots, black boots, thigh boots, gloves, black gloves, long sleeves Description
FAQ
Comments (6)
Hi! I rly love this model of yours. I've recently started creating also my own character loras, may i ask on how did you get those images on your model showcase? I also want to generate it in that style and not looking like the ones on my training images which are mostly anime screencaps. Thanks!
@ronaldmclovin92604 The final quality of Lora depends on the variation of images you have in the dataset. I believe that by using many screencaps, consequently the final quality when generating the image will look more like this.
@scarxzys_ย thanks for the great advice, i was wondering what external generator did you use to achieve those showcased images?
@scarxzys_ย how many images and what types of images did you have on your data set for this? If you could also share your training params for achieving this quality when you trained it if that would be ok ๐๐
@ronaldmclovin92604ย Stable Diffusion local on my PC
@ronaldmclovin92604ย You can collect images from zerochan, danbooru, gelbooru and safebooru, as for the parameters, I think everyone has their own ๐
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.