I present to you the legendary Meme-chanLoRA is designed for use with <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749?modelVersionId=290640">ponyDiffusionV6XL </a>or mix, for example <a target="_blank" rel="ugc" href="https://civarchive.com/models/288584?modelVersionId=324524">AutismMix SDXL</a>I couldn't create a version for Animagine in normal quality. Most likely I won't be able to release it.Model: <a target="_blank" rel="ugc" href="https://civarchive.com/models/257749?modelVersionId=290640">ponyDiffusionV6XL_v6.safetensors</a>Dataset size: 95Clip skip: 1dim32, alpha16lr=0.0003, constant, adafactorOptimizer extra arguments: scale_parameter=False relative_step=False warmup_init=FalseMy tips for generating on SDXL for AUTOMATIC1111Resolution:<ul><li>1024 x 1024</li><li>1152 x 896 or 896 x 1152</li><li>1216 x 832 or 832 x 1216</li></ul>Don't use Hires. fix. Send the picture to img2img, set CFG 0.15-0.25, then select “SD upscale” in the script, you also need to select 4x-AnimeSharp (find out how to install it) or what you like best from upscale. All other settings are default. Click generate.For me 12gb VRAM is enough for this using --xformers

Files with SHA256: 8e7381b300da7fcdca0d74baed7c603bb6af3cad92a07f84508e1e03b702a0e4