Supports 2D, 2.5D, anime, semi-realistic and other styles, with more consistent performance and enhanced prompt comprehension, plus boosted support for Danbooru artist styles.Recommended ParamsSteps: 20-40 | CFG Scale: 4-7 | Samplers: dpmpp_sde / Euler a / dpmpp2m_sde | Schedulers: beat / karras / sgm_uniform / exponentialUpscaler: esrgan_4x / 4x-UtraSharp | AD: yolov8n_v2.ptVAE is integrated in this model.My go-to settings for it: 25 steps, 6 CFG Scale, dpmpp_sde + beat scheduler, 832x1216 resolution.This model was initially built to upgrade my <a target="_blank" rel="noopener noreferrer nofollow" href="https://tensor.art/zh/models/890151544530445245/illust-Anime-2D_IL-v1">illust-Anime-2D_IL</a> checkpoint, but it’s undergone extensive modifications. While it shares some similarities with the original, there’s no direct iterative connection between them – so I’m releasing it as a standalone model.Well-crafted prompts and proper Danbooru tags are the keys to unlocking the full potential of Illustrious and NOOB. If you like this model, your feedback and post shares would be much appreciated! Hope you enjoy using it!

illust-Anime-2D Ω IL

Steps: 20-40, CFG scale: 4-7, Sampler: dpmpp_sde_gpu, euler，euler a，scheduler: beta，normal，simple，

Files with SHA256: daa8b0f71ef9bcb5ff2abe2ef7877d2db14dc1d7119f076792bc579229e957c5