Update README.md
Browse files
README.md
CHANGED
|
@@ -10,4 +10,50 @@ tags:
|
|
| 10 |
base_model:
|
| 11 |
- Tongyi-MAI/Z-Image-Turbo
|
| 12 |
base_model_relation: quantized
|
| 13 |
-
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 10 |
base_model:
|
| 11 |
- Tongyi-MAI/Z-Image-Turbo
|
| 12 |
base_model_relation: quantized
|
| 13 |
+
---
|
| 14 |
+
For more information (including how to compress models yourself), check out https://huggingface.co/DFloat11 and https://github.com/LeanModels/DFloat11
|
| 15 |
+
|
| 16 |
+
Feel free to request for other models for compression as well, although models whose architecture I am unfamiliar with might be slightly tricky for me.
|
| 17 |
+
|
| 18 |
+
### How to Use
|
| 19 |
+
|
| 20 |
+
#### ComfyUI
|
| 21 |
+
Install my own fork of the DF11 ComfyUI custom node: https://github.com/mingyi456/ComfyUI-DFloat11-Extended. After installing the DF11 custom node, use the provided workflow [json](z_image_turbo_bf16-DF11-workflow.json), or simply replace the "Load Checkpoint" node of an existing workflow with the "Load Checkpoint with DFloat11 Unet" node. If you run into any issues, feel free to leave a comment. The workflow is also embedded in the below [png](z_image_turbo_bf16-DF11-workflow.png) image.
|
| 22 |
+
|
| 23 |
+
|
| 24 |
+
|
| 25 |
+
#### `diffusers`
|
| 26 |
+
Refer to this [model](https://huggingface.co/mingyi456/Z-Image-Turbo-DF11) instead.
|
| 27 |
+
|
| 28 |
+
### Compression Details
|
| 29 |
+
|
| 30 |
+
This is the `pattern_dict` for compressing Z-Image-based models in ComfyUI:
|
| 31 |
+
|
| 32 |
+
```python
|
| 33 |
+
pattern_dict_comfyui = {
|
| 34 |
+
r"noise_refiner\.\d+": (
|
| 35 |
+
"attention.qkv",
|
| 36 |
+
"attention.out",
|
| 37 |
+
"feed_forward.w1",
|
| 38 |
+
"feed_forward.w2",
|
| 39 |
+
"feed_forward.w3",
|
| 40 |
+
"adaLN_modulation.0"
|
| 41 |
+
),
|
| 42 |
+
r"context_refiner\.\d+": (
|
| 43 |
+
"attention.qkv",
|
| 44 |
+
"attention.out",
|
| 45 |
+
"feed_forward.w1",
|
| 46 |
+
"feed_forward.w2",
|
| 47 |
+
"feed_forward.w3",
|
| 48 |
+
),
|
| 49 |
+
r"layers\.\d+": (
|
| 50 |
+
"attention.qkv",
|
| 51 |
+
"attention.out",
|
| 52 |
+
"feed_forward.w1",
|
| 53 |
+
"feed_forward.w2",
|
| 54 |
+
"feed_forward.w3",
|
| 55 |
+
"adaLN_modulation.0"
|
| 56 |
+
)
|
| 57 |
+
}
|
| 58 |
+
```
|
| 59 |
+
|