About v2.0
This new version is using a brand new unique way of creating Lora's.
This result in a better contrast, likeness, flexibility and morphology while using half the size of my traditional Lora training. In this case it's 3.5x less
Use a Lora strength between 0.85 to 1.0
Disclaimer: This artificial intelligence model ("the Model") is developed solely for artistic expression and technological demonstration, not for creating or distributing illicit or inappropriate content. The creator expressly disapproves of such use. By downloading or using the Model, you ("the User") agree not to misuse it in any manner contrary to these intentions and accept all responsibility for compliance with these terms.
Description
This is the slimed down of the big model. It seems to have lost a bit of quality but I cannot compared 1 to 1 since it produce different output than the full model.
FAQ
Comments (6)
Amazing results!
You may not want to reveal all your tricks which is understandable, but any advise on settings/ things you learned from your experiment?
Won't reusing "ohwx woman" as a keyword introduce ambiguity when we try to use two or more of your character loras together in the same image?
You shouldn't have labeled the first pic 13+ -- it's not, and that causes it to be blurred out when searching. Not sure if it's too late to fix it. Cheers.
You can also switch from ohwx woman - (ohwx woman) - ((ohwx woman)) to increase the likeness,should i put this to positive prompt?anyway,when i train dataset of my own pics,the lora result image is not realistic and not exactly like of my own profile,can u share your config file?
Love This Lora, I'm getting so much better results from this new version. Thanks! I'm looking forward to an updated Jennifer Lawrence!
Not working for me?
Looks like we don't have an active mirror for this file right now.
CivArchive is a community-maintained index — we catalog mirrors that volunteers upload to HuggingFace, torrents, and other public hosts. Looks like no one has uploaded a copy of this file yet.
Some files do get recovered over time through contributions. If you're looking for this one, feel free to ask in Discord, or help preserve it if you have a copy.