character LoRA for D.W Read
Description
FAQ
Comments (5)
It likes to generate images of her facing away from me, I don't know what that's about. Otherwise it mostly works.
There's a like 2-3 images of her from behind in the training data. could add "from behind" to negatives or just prompt it a facial expression or directly tell it "facing viewer"
"facing viewer" also generates her facing away, which is odd. And so does putting "from behind" into the negative prompt, though one less than last time.
@tombot just curious, what checkpoint and extra loras are you using? i can't seem to replicate it.
@foxai143 I'm using the on-site image generator, this Lora and the basic PonyXL model, thats about it. I'm also using the prompt you supplied as well as the prompts you suggested in the comment and some added stuff: "standing", "solo" and "garden background". Though this happens even when I didn't add those extra prompts.
Details
Files
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.