Add automated datasheet
Browse files
README.md
ADDED
|
@@ -0,0 +1,37 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: other
|
| 3 |
+
library_name: diffusers
|
| 4 |
+
tags:
|
| 5 |
+
- text-to-image
|
| 6 |
+
- stable-diffusion-xl
|
| 7 |
+
- diffusers
|
| 8 |
+
- slerp
|
| 9 |
+
- merge
|
| 10 |
+
base_model:
|
| 11 |
+
- NoobAI-XL-Vpred
|
| 12 |
+
- IllumiYume-XL-Vpred
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
# SpacesFusion_V1
|
| 16 |
+
|
| 17 |
+
**SpacesFusion_V1** is a high-precision SLERP merge focused on preserving the superior anatomy of NoobAI XL while injecting the high-resolution capabilities and newer dataset knowledge (2025 tags) from IllumiYume V3.5.
|
| 18 |
+
|
| 19 |
+
## Composition Strategy (The "Nuclear" Recipe)
|
| 20 |
+
|
| 21 |
+
This model was merged using a specific block-weighted SLERP technique to ensure V-Prediction stability.
|
| 22 |
+
|
| 23 |
+
* **Base Architecture:** NoobAI XL v1.0 (Dominant Structure)
|
| 24 |
+
* **Detail Injector:** IllumiYume V3.5 (Dataset & Texture)
|
| 25 |
+
|
| 26 |
+
### Block Weights
|
| 27 |
+
* **Input Blocks (Anatomy/Structure):** Highly biased towards NoobAI to prevent anatomical degradation.
|
| 28 |
+
* **Middle Blocks:** Balanced mix.
|
| 29 |
+
* **Output Blocks (Texture/Lighting):** Progressive injection of IllumiYume parameters to allow for higher resolution scaling (1536px+) and newer art style rendering.
|
| 30 |
+
|
| 31 |
+
## Recommended Settings
|
| 32 |
+
* **Resolution:** 1024x1024 up to 1536x1536 (Native Support)
|
| 33 |
+
* **CFG:** 5-7
|
| 34 |
+
* **Sampler:** Euler a / DPM++ 2M SDE Karras
|
| 35 |
+
* **V-Prediction:** ENABLED (Critical)
|
| 36 |
+
|
| 37 |
+
Created by **Bl4ckSpaces** using Automated Evolutionary SLERP.
|