clip-interrogator-ext
clip-interrogator-ext copied to clipboard
Documentation
Hi, is there any kind of documentation to the extension/difference between the models/suggestive usage ? TIA :)
in the extension if you click the about tab it has recommended CLIP model to choose there. basically if you are generating images with a stable diffusion 1.X model then you want to use ViT/L and if you are using a stable diffusion 2.X model then you want to use ViT-H.
thanks!
Could there be a hint added for which model to use for SDXL?