Trained on Anything V5, 15 images.
Works great with ThorneMix_v1, Dark Sushi 2.5D 大颗寿司2.5D, darkSushi 2.5D_V2.0, Coffee NSFW, Yotta Mix v5.
GrahamAingPikas-ClearVAE 2.3 also produces good results, tho not as good as those above ( at least to my preference).
I use Euler A
23 sampling steps ( I use this little because of testing, feel free to use more)
7 CFG
Thumbnail Model: ThorneMix_v1