Raehoshi illust XL

Raehoshi Illust XLan enhanced iteration built upon the Illustrious XL v01 model. It aims to elevate the visual style by addressing some of the limitations in the original, such as oversaturation and artifact noise. While these issues are not entirely eliminated, noticeable improvements have been made. The goal is to deliver a more polished, balanced output while staying true to the strengths of the base model. Enjoying my work? Consider supporting the development of my models by buying me a coffee on <a href="https://ko-fi.com/raelina" target="_blank">Ko-fi</a> Also check out my other models on Mage: <a href="https://www.mage.space/play/55ee9a305fc7e3572509def13acbefbf" target="_blank">Raepix Focus XL</a> / <a href="https://www.mage.space/play/08df2e9b11b4d9587dcbb6846a9dda11" target="_blank">Raepix Focus XL Lightning</a> / <a href="https://www.mage.space/play/41658a734e3918e89e700c76f103ffc7" target="_blank">Raevitamix Realistic</a> / <a href="https://www.mage.space/play/b1586a8de40e97411fdcda1fe03ceb60" target="_blank">Raevista XL</a> Recommended settingPositive prompt :masterpiece, best quality, very aesthetic,*Special tags: dramatic lightingNegative prompt :lowres, bad quality, worst quality, very displeasing, bad anatomy, sketch, jpeg artifacts, signature, watermark,Steps : 28CFG : 5-6Sampler : euler_aSupported resolution :1024 x 1024, 1152 x 896, 896 x 1152, 1216 x 832, 832 x 1216, 1344 x 768, 768 x 1344, 1536 x 640, 640 x 1536Hires.fix Setting:Upscaler : 4x NMKD YandereNeo XLHires step : 10-15Denoise : 0.1-0.3 Training DetailsThe model was developed using a two-stage fine-tuning process. In Stage 1, new series and characters were introduced into the model. Stage 2 focused on fixing issues and enhancing the overall style for improved output.Stage 1Dataset : v1-31k and v2-37kHardware : 2x A100 80gbBatch size : 32Gradient accumulation steps : 2Learning rate : 6e-6Text encoder : 3e-6Epoch : 15Stage 2Dataset : v1-2.5k and v2-2.3kHardware : 1x A100 80gbBatch size : 48Gradient accumulation steps : 1Learning rate : 3e-6Text encoder : disableEpoch : 15 List of New Series/Characters Trained:zenless zone zero series characterwuthering waves series characterhololive from justice and dev isHSR character from 2.2 - 2.5 versiontokidoki bosotto roshia-go de dereru tonari no alya-san seriesDandadanFor character prompts, please refer to the Danbooru site for accurate tags and references. License<a href="https://freedevproject.org/faipl-1.0-sd/" target="_blank">Fair AI Public License 1.0-SD</a> Special thanks to Joe for supporting my works

Files with SHA256: 4fd374bc2158ef10c639272afbf6775f0eb3e61464850a2e5cf0ba1740ad4e0d