Downloads · 30 days
17
4% of all-time downloads
Ilustrious-Nova-V4.0-PRO
Bl4ckSpaces/Ilustrious-Nova-V4.0-PRO
Ilustrious-Nova-V4.0-PRO is a text-to-image model from Bl4ckSpaces. Use it when you need an image from a text prompt. It is set up for diffusers. The card lists the license as openrail++.
V4.0 PRO addresses the high-contrast sensitivity of the original V4.0 logic by implementing Cross-Reference Rescaling. It combines the superior anatomy of IllumiYume with the disciplined color range of NoobAI.
All-time downloads
412
Public
Repo size
6.9 GB
Likes
1
Public
Repo makeup
Click a slice to open those files.
.safetensors6.9 GB · 100%
At a glance
- Task
- Text-to-Image
- Library
- diffusers
- License
- openrail++
- Access
- Public
- Created
- Jan 26, 2026
- Updated
- Jan 26, 2026
- SHA
- 3b96724e
Model card
From the Hugging Face model README
Ilustrious-Nova-V4.0-PRO (The Masterpiece)
V4.0 PRO addresses the high-contrast sensitivity of the original V4.0 logic by implementing Cross-Reference Rescaling. It combines the superior anatomy of IllumiYume with the disciplined color range of NoobAI.
The "Cooling" Technology
The main issue with IllumiYume (Base) is its tendency to "burn" or over-saturate at high CFG levels. V4.0 PRO fixes this by splitting the normalization targets:
- Anatomy Layers (Input/Middle): Rescaled to IllumiYume.
- Why? To maintain the rigid skeletal structure and perfect fingers of the base model.
- Color/Style Layers (Output): Rescaled to NoobAI.
- Why? Instead of inheriting IllumiYume's high contrast, we force the output colors to match NoobAI's softer, wider dynamic range. This effectively "tames" the burn.
- Safety Dampener: A 0.98x multiplier is applied to output blocks to prevent pixel clipping at high CFG (8.0+).
Recommended Settings
- Sampler: Euler a / DPM++ 2M SDE
- CFG Scale: 5.0 - 9.0 (Fully safe at high CFG!)
- Steps: 28+
- V-Prediction: ENABLED
- Task
- Text-to-Image
- Library
- diffusers
- License
- openrail++
- Created
- Jan 26, 2026
- Updated
- Jan 26, 2026