different image quality with different tools but same settings
Hello Lykon!
Thank you for sharing the amazing model. I tested your model in Leonardo.ai to do image-to-image generation, and the results were excellent.
However, when I want to use Diffuser with the DreamShaper v5 to generate the same output with the same prompts, input image, and other settings (seeds, steps ), the image quality worsens. I know Leonardo.ai uses its own Leonardo sampler, so the output may differ. However, there's a noticeable gap between them which makes me confused. I'm using Tesla M60 with a 16G GPU to do the generation.
Do you know what are the possibilities that make the changes? Thank you very much!
No idea, I've never used it on leonardo.ai.
Try to compare your results with the images I posted here or on civitai, since those are the ones I generated myself.