Trained with <a rel="ugc" href="https://huggingface.co/WarriorMama777/OrangeMixs#abyssorangemix3-aom3">AOM3A1</a>All samples are not fixed with i2i and just got upscaled with the settings below.Good result with<ul><li>Model : AOM3A1</li><li>VAE : kl-f8-anime2</li><li>LoRA Weight : around 0.7</li><li>Sampling method : DPM++ SDE Karras</li><li>Sampling Step : 30~40</li><li>CFG Scale : around 9~10</li><li>Upscaler : R-ESRGAN 4x+ Anime6B</li><li>Denoising : around 0.5</li></ul>

Files with SHA256: 062c8e2ff1e966bed16fa8fff2e33dfd302d53777d28de5b57bfe96e290a9719