I have no idea how people are sleeping on these. I implemented your power noise ksampler workflow into an img2img finetune / dreambooth workflow. This completely changed things for me and it looks exactly like the person. It has been one of the biggest single enhancements I've seen in a long time. Especially when adding perturbed guidance and automatic cfg with your workflows blend settings.
The only issue I have is with the stable video diffusion power noise ksampler. I wanted to hook it up but it's not quite working right. One time I got the most amazing SVD video with this ksampler I've ever seen, so I know it's possible. but I lost the power noise sampler settings and workflow. Is there any chance you could share an SVD workflow for SVD using this? I am going to spread the word because this is great.
Currently implementing this into COSXL workflow, and converted my sdxl finetune to COSXL to subtract out SDXL base model and clip. Will see how it turns out.
I have no idea how people are sleeping on these. I implemented your power noise ksampler workflow into an img2img finetune / dreambooth workflow. This completely changed things for me and it looks exactly like the person. It has been one of the biggest single enhancements I've seen in a long time. Especially when adding perturbed guidance and automatic cfg with your workflows blend settings.
The only issue I have is with the stable video diffusion power noise ksampler. I wanted to hook it up but it's not quite working right. One time I got the most amazing SVD video with this ksampler I've ever seen, so I know it's possible. but I lost the power noise sampler settings and workflow. Is there any chance you could share an SVD workflow for SVD using this? I am going to spread the word because this is great.
Currently implementing this into COSXL workflow, and converted my sdxl finetune to COSXL to subtract out SDXL base model and clip. Will see how it turns out.