Downloads · 30 days
25
22% of all-time downloads
Ilustrious-Nova-V3.5-Safe-DARE
Bl4ckSpaces/Ilustrious-Nova-V3.5-Safe-DARE
Ilustrious-Nova-V3.5-Safe-DARE 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++.
This is the STABILIZED version of V3.5. It fixes the "white screen/static noise" issue by implementing tensor clamping and reduced rescale factors.
All-time downloads
113
Public
Repo size
6.9 GB
Likes
0
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 25, 2026
- Updated
- Jan 25, 2026
- SHA
- d66cabdd
Model card
From the Hugging Face model README
Ilustrious-Nova-V3.5-Safe-DARE (V3.5 Safe-DARE)
This is the STABILIZED version of V3.5. It fixes the "white screen/static noise" issue by implementing tensor clamping and reduced rescale factors.
The Recipe (Inverse Merge)
- BASE (Style): NoobAI XL v1.0
- DONOR (Anatomy): IllumiYume v3.5
The "Safe-DARE" Method
We injected the structural data (Input Blocks) from IllumiYume into NoobAI using DARE, but with Safety Clamping (-8 to 8) and Limited Rescaling (Max 4x) to prevent gradient explosion.
Recommended Settings
- Sampler: Euler a / DPM++ 2M SDE
- CFG: 5.0 - 7.0
- Steps: 28+
- V-Prediction: ENABLED
- Task
- Text-to-Image
- Library
- diffusers
- License
- openrail++
- Created
- Jan 25, 2026
- Updated
- Jan 25, 2026