[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/vt/ - Virtual Youtubers

Search:

View post   
  [?]In ghost mode, only threads with non-archived posts will be shown
>> No.40089468 [View]
File: 2.84 MB, 2501x4000, 00071-1148873118-best quality, solo, hyakumantenbara salome, virtual youtuber, (x_1.0).jpg [View same] [iqdb] [saucenao] [google]
40089468

>>40088580
Updated HLL model.
Now it should know nijis and indies better, but you'll still need to specify details like hair and eye color.
Also added a lot of artists, nsfw tags, and other random stuff.
Tag statistics: https://litter.catbox.moe/9aoygs.txt - a lot of weird stuff in there, not everything works
simple NSFW examples: https://litter.catbox.moe/m6whar.jpg

Download link: pixeldrain com/u/8kfyU5C3

I recommend to merge with AnythingV3+Elysium or nutmegmixv2, and using nai VAE.
Also try adding "comic, monochrome, greyscale, speech bubble, guro, censored, bar censor" to usual negatives.

Also, I'm still planning to train a different, better model in mid-january.

Navigation
View posts[+24][+48][+96]