This LoRA was trained on still images of bizarrely distorted faces. I t responds to the word caricature and descriptions of exaggerated facial characteristics.
Modellquelle
Quellenbeschreibung
This LoRA was trained on still images of bizarrely distorted faces. I t responds to the word caricature and descriptions of exaggerated facial characteristics.
An example of a prompt that will draw out all the exaggerated features is: caricature of a person with exaggerated features including a long face, enormous bulging eyes, big drooping ears, a bulbous nose, gigantic lips, and an expressive face.
Krea 2:
Krea 2 gives much more diverse head shapes and facial expressions! All samples were created with er_sde/simple.
Z-Image Turbo:
All sample images were created with 9 steps using res_multistep/simple in Comfyui. Doesn't seem to work on animals as powerfully as the Wan version does.
Quellen
1 QuelleModellartefakte
1 ArtefaktGoofballs - WAN-21-22.safetensors
safetensors · 293 MB · SHA-256 a1a644da28b4…3559 · Civitai
HerunterladenQuellenauszüge
1 AuszugSource context: 895 downloads · Type LORA · Base model Krea 2
Wan:
The largest portrait orientation I could go without ridiculous mutations was 768x1024 . Hands are usually borked, unless you do some kind of hires-fix. Also, each gallery image contains a ComfyUI workflow where you can see the settings used.