Superfacialized v.1.3
Works best with photorealistic models based on SD-1.5
Best results with VAE: vae-ft-mse-840000-ema-pruned
Samples using the model: Hrl32, thanks!!!
----
Trigger word "facializ3d" to tune "facials" and "bukkake" tokens. Good results with with "0.7" strenght <lora:superfacialized-v1.3:0.7>
Example:
a photorealistic photo portrait of facializ3d happy woman at the bathroom, facials, bukkake <lora:superfacialized-v1.3_3107:0.7>
Negative promt:
3d, render, painting, anime, manga, illustration, (deformed, distorted, disfigured:1.3), poorly drawn, bad anatomy, wrong anatomy, extra limb, missing limb, floating limbs, (mutated hands and fingers:1.4), disconnected limbs, mutation, mutated, ugly, disgusting, blurry, amputation, blue eyes, bigeyes, teeth, piercing
Description
Init version
FAQ
Comments (9)
Am I doing something wrong? This lora specifically just does not do anything to my gens. I've tried redownloading it. I've tried removing basically everything from the prompt except the trigger words and a face for them to affect. I turn the strength all the way up, put it in (((triple))), Nothing. Not a drop of goo. I'm doing something wrong.
anyone else getting an error running this in automatic1111?
RuntimeError: output with shape [4, 320, 1, 1] doesn't match the broadcast shape [4, 320, 3, 3]
Doesn't work. I use same model or any 1.5 ver models, use right tags and wae.
Works great! These prompts help ((facials)),bukkake(cum on face),blonde hair, cum-covered, ((cum-dripping))
doesn't work
Not sure why it doesnt work, other loras work fine, i used the prompts correctly, even copied the ones used here, so its not a problem on my end, seems to be an issue for other people too
Use sd-webui-additional-networks extension to get this lora to work
works for me but if you're copying and pasting the example be cautious that there is a typo should be "superfacialized-v1.3:0.7" not "superfacialized-v1.3_3107:0.7" It worked once I changed this.
im using zavant.ai and they are coming out pretty good, but at like the 50% generation mark it totally changes the image and i notice that i like the image more up to that point then it changes, any thoughts on why thats happening and which parameter to play around with??
Details
Files
superfacialized-v1.3.safetensors
Mirrors
superfacialized-v1.3.safetensors
superfacialized-v1.3.safetensors
superfacialized-v1.3.safetensors
superfacialized_v13.safetensors
Superfacialized.safetensors
superfacialized_v13.safetensors
superfacialized-v1.3.safetensors
superfacialized-v1.3.safetensors
superfacialized-v1.3.safetensors
superfacialized_v13.safetensors
Available On (3 platforms)
Same model published on other platforms. May have additional downloads or version variants.