Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,19 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model_relation: quantized
|
| 3 |
+
tags:
|
| 4 |
+
- dfloat11
|
| 5 |
+
- nvfp4
|
| 6 |
+
- df11
|
| 7 |
+
- lossless compression
|
| 8 |
+
- 70% size, 100% accuracy
|
| 9 |
+
language:
|
| 10 |
+
- en
|
| 11 |
+
base_model:
|
| 12 |
+
- black-forest-labs/FLUX.2-klein-4B
|
| 13 |
+
---
|
| 14 |
+
|
| 15 |
+
Just some testing.
|
| 16 |
+
|
| 17 |
+
The bf16 layers are compressed using the Dfloat11 lossless compression mixed with nvfp4 layers.
|
| 18 |
+
|
| 19 |
+
This isnt the perfect balance between nvfp4 layers and Dfloat11 compressed layers and changes a good amount from model to model but it is a start.
|