Kandinskiy 2.2
About
This model enhances text-to-image generation with advanced encoders and ControlNet, delivering aesthetic, accurate images. It leverages a CLIP-ViT-G encoder for improved text comprehension and supports iterative control mechanics, expanding possibilities in image manipulation.
Percs
High-resolution output
Multilingual support
Advanced editing tools
Fast generation
Settings
Negative Prompt- Things you don't want to see in the output
Width-
Height- undefined
Inference Steps- Number of denoising steps
Inference Steps Prior- Number of inference steps for prior