Self Attention Guidance: Improve image background (ComfyUI)

Self-attention Guidance (SAG) enhances details in an image while preserving the overall composition. It is useful for fixing nonsensual details that give the AI image away.
ComfyUI workflow download (Go to comfyUI section)
stable-diffusi...
Stable Diffusion course
stable-diffusi...

[@superlucky4499]

Thank you for the explanation! Please do more!

[@integragemi7234]

for loras the node goes before the lora or after the lora

[@SejalDatta-l9u]

Thank you- what if you load a reference image?

[@paganiriccardo]

What about if I add a LoRA? SAG node goes before or after the lora loader?

[@stable-diffusion-art]

Good question. It can be before or after. The SAG node merely set the callback function to be used during sampling.

[@paganiriccardo]

@stable-diffusion-art thanks for your answer! So, staying at an intuitive level, I guess I would have two slightly different results, but both improved by the SAG algorithm. I will do some tests, thanks again