V2 has more images and is just an overall improvement. Faces are better but still not perfect. It shouldn't give any accidental comic elements.

Works well so far, especially for my second attempt ever at making a LoRa. This is trained to reproduce styles and male anatomy. I did not use any female oriented images, and don't care to know if it can produce them. Same goes for fat fetishes. The artist does plenty of both. This was also trained exclusively on SFW images, as this is an SFW artist mostly, so NSFW will take a few seconds longer and may produce some... odd censoring depending on prompts.

Struggles with faces sometimes, will often add background characters, and also often makes images into comics. I should be able to fix all of that in a new version pretty easily though. For now, it's still a great one that can do a lot of things. It's especially good at duos, busts, and dynamic poses. *This is the version where that is mostly fixed.

I will take this down if the artist objects.

Support them here at Userpage of Ronnie Sparco -- Fur Affinity [dot] net

P.S. try to add "lens flare" to your prompts when using this, plenty of images it was trained on used it and is just a neat effect.