<h1 id="rouwei-from-hades-v0.2">🌸 Rouwei from HaDeS v0.2</h1><h3 id="what's-new-in-v0.2">⚙️ What’s new in v0.2</h3><ul><li>✨ Improved overall image quality — sharper focus, cleaner details, and smoother rendering.</li><li>🌌 Enhanced backgrounds — better depth, color harmony, and structure.</li><li>🔵 Blue tone nearly removed — natural color balance restored.</li><li>🎨 Closer alignment to Illustrious from HaDeS aesthetic — softer light, cohesive shading, and finer tonal range.</li><li>⚡ More stable composition across complex prompts and long descriptions.</li></ul><hr /><h3></h3><h1 id="rouwei-from-hades-v0.1">🌸 Rouwei from HaDeS v0.1</h1>New RouWei-based finetune trained on 0.76M images. 🔮 Best prompt adherence among SDXL anime models. 🎨 <a target="_blank" rel="ugc" href="https://drive.google.com/file/d/1QMULgKmiyX1PZKGMblUs41IFsgKS5AEX/view?usp=drive_link">50,000+ artists &amp; styles</a> with vibrant colors, smooth gradients without watermarks. 🖌️ Works with tags + natural text (hybrid captions). ⚡ Flexible: short or long prompts.Sampling: Euler, Euler A - 20–28 steps · CFG 3–5 Quality tags: <code>masterpiece, best quality</code> / <code>low quality, worst quality</code>👉 Separate artist/style tags with <code>BREAK</code> (or put them at the end). ❤️ Share your gens &amp; support future training!

RouWei from HaDeS

Rouwei from HaDeS v0.1 Large-scale RouWei retraining. ✅ Best prompt adherence (anime SDXL) ✅ 0.76M images, 50k+ artists/styles ✅ No watermarks, vibrant colors, smooth gradients ✅ Tags + natural text, full flexibility⚙️ Euler_a · 28 steps · CFG 3–5

Files with SHA256: 3e61d3ca3b5912469b3d0bcf00a3e1df7b22d00ded4e5303f84703ada14a9946