Was kind of unhappy with one of the better Charlie Barkin LoRAs was restricted and the others were not as good. So here's my attempt. Not as good, but good enough.
Description
Comments (2)
That is a lot of neck. Was most the dataset headshots or shoulders up shots?
I'd say 50 face-focused, 50 full body. I was avoiding Non-anthro shots/on fours for the most part since CivitAI seems to flag anything feral as restricted content. The one LoRA I liked was flagged due to it, so why I made it. Also more than 500 images to train becomes restricted in generations unless you go about .50 strength.
Also while I could find some Charlie shots otherwise, alot were with other characters. I didn't want to muddle the dataset with multiple characters, even partially because they could artifact into the background.
Edit: also I try to restrict what I use for the main generations to WAI-Illustrious, and reduce the number of LoRAs used because I want to show a more "Out of the Box" results vs off-site generations of extra LoRA usage.
Details
Available On (1 platform)
Same model published on other platforms. May have additional downloads or version variants.