on I did it ... (starting to get results with SDXL) · c/stable_diffusion · 2 pts · 3yThanks for the info 😀, it explains some of the weirds results I got with my first trials... This page is full of useful info, like the best CFG scale for SDXL: "SDXL models respond well to a wider range (IE: 4-12)"
Thanks for the info 😀, it explains some of the weirds results I got with my first trials...
This page is full of useful info, like the best CFG scale for SDXL: "SDXL models respond well to a wider range (IE: 4-12)"