This model is an fp8 model, so we have not tested it with anything other than ComfyUI.

This model is a model that has been heavily adjusted by mixing animagine, pony, illustrious, and Noob in a ratio of 0.1 after personal learning of SDXL1.0. LoRA is effective for all models: animagine, pony, illustrious, and Noob. (Effectiveness varies for each model.)

refer to

negativeprompt: (low quality, bad quality), bad anatomy, oldest, username, text, nsfw, noise,

clipskip-2 28step cfg6 dpmpp_2m_sde karras