Model Information






Description
🌀 Noob from HaDeS — v2 Series
✨ Cinematic Stylization · Realistic Lighting · Enhanced 2.5D Depth
Noob from HaDeS v2 refines the balance between illustrated detail and environmental realism, establishing a polished 2.5D aesthetic with cinematic emotion.
It renders characters with near-tangible presence, illuminated by natural light and seamlessly integrated into realistic settings.
✅ Key Visual Traits:
🎨 2.5D Stylization Core — merges sharp illustrative precision with volumetric shading and lifelike textures reminiscent of digital photography.
🌤️ Realistic Lighting System — global light behavior and natural reflections; skin, fabrics, and materials react coherently to ambient conditions.
💎 Cinematic Color Tone — vibrant yet organic palettes, smooth contrast transitions, and nuanced skin tones.
🪞 Material Precision — expertly handles transparency, moisture, reflections, and complex fabrics without breaking realism or sharpness.
🧠 Adaptive Rendering — reacts fluidly to a wide range of prompts, from intimate portraits to full scenic compositions, while maintaining aesthetic harmony.
⚙️ Recommended Settings (v2):
Sampler: Euler / Euler a
Steps: 28–35
CFG Scale: 3-5
Resolution: 1152×1536 or 1024x1536 or SDXL resolutions
VAE: baked
💡 Notes:
Perfect for creators aiming for 2.5D expressive characters over realistic backgrounds, with polished and cohesive lighting control.
— DeViLDoNia / HaDeS
🌀 Noob from HaDeS — v1 Series
✨ First Release · Noob AI Core · HaDeS Signature Refinement
Noob from HaDeS v1 marks the beginning of a brand-new branch in the illustrious from HaDeS lineup.
Unlike Illustrious from HaDeS, which was built on Illustrious XL, this model is based on Noob AI — adapted, refined, and reshaped with the HaDeS touch.
✅ What’s new in Noob from HaDeS v1:
🔀 Style Variety — inherits Noob AI’s ability to shift between multiple artistic styles, from clean stylized illustration to semi-realistic scenes.
🖌️ Flexible Prompt Adherence — works reliably with both simple and complex prompts, producing diverse yet controlled outputs.
🎛️ Stability & Coherence — tuned for consistent generations, balancing detail with creative variety.
⚡ Foundation for a New Line — this first release sets the stage for future updates, combining Noob AI’s freedom with the refined direction of the HaDeS aesthetic.
🔁 Recommended Settings (v1):
Sampler: Euler, Euler a (best results) or DPM++ 2M for more variety
Steps: 25–30
Resolution: 1024×1536 / 1152×1536
CFG Scale: 5–7
VAE: baked
💡 Notes:
Noob from HaDeS is not a replacement for Illustrious from HaDeS — it’s a parallel path, built to explore new stylistic possibilities while keeping the HaDeS identity.
v1 is stable but experimental: expect consistent results even with minimal tweaking.
Future updates will further refine anatomy, lighting, and stylistic specialization.
— DeViLDoNia / HaDeS
NooB from HaDeS
v2
Model Details
- Type
- AI Model
- Task
- text-to-image
- Subtype
- Safetensors / Checkpoint AI Model
- Created
- Updated
- May 27, 2026