Version 1, might make an updated version later with more fluidity in poses.


Just a quick one, trained on sd 1.5 checkpoint. Works best with realistic checkpoints like EdgeOfRealism, Level4, RealisticVision,... but also works fine in Dreamshaper and RevAnimation as far as I have tested.
Recommended value between 0.5 and 0.8, depending on checkpoint. I did notice some rigidness in poses, for which I apologize in advance. As for training, I think I did something like 10 epochs 20 repeats for a total of 2500 steps.
Again, I am pretty new at this, so feedback is welcome!
Version 1, might make an updated version later with more fluidity in poses.
1. Os direitos dos modelos repostados pertencem aos criadores originais.
2. Criadores originais que desejam reivindicar seus modelos devem contatar a equipe do SeaArt AI pelos canais oficiais. Clique para reivindicar
