<h3 id="about-omegawaifu">About OmegaWaifu</h3>Hey! Thanks for checking out OmegaWaifu.This isn't a standard fine-tune—it's a custom merge of several different LoRAs custom trained. I created the datasets and blended their strengths to create a versatile model with a clean, vibrant, albeit spicy, anime style.It’s designed to take the best features from each LoRA to make generating beautiful characters easy.Give it a try and see what you can create<h3 id="quick-settings-guide">Quick Settings Guide</h3>Base Generation<ul><li>Sampler: DPM++ 2M Karras, Euler, or Euler A</li><li>Steps: 20 - 40</li><li>CFG Scale: 7</li><li>Resolution: 512x640 or 576x768</li></ul>Hires. Fix (Highly Recommended!)You'll want to use Hires. Fix to get that sharp, high-quality look.<ul><li>Upscaler: Latent</li><li>Denoising Strength: 0.5 - 0.7<ul><li>(Use 0.5 for a softer look, or push it toward 0.7 for more fine details.)</li></ul></li><li>Upscale by: 1.5x to 1.8x</li></ul>

Files with SHA256: 3f5268bab8f12fe35df21f0f1a28c8c21552a09b7521a4027d149bf7b8883b89