Update README.md
Browse files
README.md
CHANGED
@@ -6,12 +6,12 @@ license: apache-2.0
|
|
6 |
|
7 |
| Converted LoRAs | Original LoRAs |
|
8 |
|-----------------|----------------|
|
9 |
-
| [
|
10 |
-
| [
|
11 |
|
12 |
|
13 |
## 🛠️ Conversion References
|
14 |
-
**Script:** [
|
15 |
|
16 |
Based on
|
17 |
- **Nunchaku Issue:** [ComfyUI-nunchaku #340](https://github.com/mit-han-lab/ComfyUI-nunchaku/issues/340)
|
@@ -22,12 +22,12 @@ Based on
|
|
22 |
## 🔄 Example Conversion Log
|
23 |
|
24 |
```bash
|
25 |
-
Running
|
26 |
|
27 |
🔄 Universal final_layer.adaLN LoRA patcher (.safetensors)
|
28 |
|
29 |
Enter path to input LoRA .safetensors file: Flux_kontext_deblur.safetensors
|
30 |
-
Enter path to save patched LoRA .safetensors file:
|
31 |
|
32 |
✅ Loaded 610 tensors from: Flux_kontext_deblur.safetensors
|
33 |
|
@@ -45,7 +45,7 @@ Enter path to save patched LoRA .safetensors file: flux_kontext_deblur_comfyui.s
|
|
45 |
- lora_unet_final_layer_adaLN_modulation_1.lora_down.weight (shape: torch.Size([16, 3072]))
|
46 |
- lora_unet_final_layer_adaLN_modulation_1.lora_up.weight (shape: torch.Size([64, 16]))
|
47 |
|
48 |
-
✅ Patched file saved to:
|
49 |
Total tensors now: 612
|
50 |
|
51 |
🔍 Verifying patched keys:
|
|
|
6 |
|
7 |
| Converted LoRAs | Original LoRAs |
|
8 |
|-----------------|----------------|
|
9 |
+
| [flux_kontext_deblur_nunchaku.safetensors](https://huggingface.co/lym00/comfyui_nunchaku_lora_patch/blob/main/flux_kontext_deblur_nunchaku.safetensors) | [Flux.1 Kontext Deblur](https://civitaiarchive.com/models/1737381) |
|
10 |
+
| [flux_kontext_face_detailer_nunchaku.safetensors](https://huggingface.co/lym00/comfyui_nunchaku_lora_patch/blob/main/flux_kontext_face_detailer_nunchaku.safetensors) | [Flux.1 Kontext Face Detailer](https://civitaiarchive.com/models/1752776) |
|
11 |
|
12 |
|
13 |
## 🛠️ Conversion References
|
14 |
+
**Script:** [patch_comfyui_nunchaku_lora.py](https://huggingface.co/lym00/comfyui_nunchaku_lora_patch/blob/main/patch_comfyui_nunchaku_lora.py)
|
15 |
|
16 |
Based on
|
17 |
- **Nunchaku Issue:** [ComfyUI-nunchaku #340](https://github.com/mit-han-lab/ComfyUI-nunchaku/issues/340)
|
|
|
22 |
## 🔄 Example Conversion Log
|
23 |
|
24 |
```bash
|
25 |
+
Running patch_comfyui_nunchaku_lora.py
|
26 |
|
27 |
🔄 Universal final_layer.adaLN LoRA patcher (.safetensors)
|
28 |
|
29 |
Enter path to input LoRA .safetensors file: Flux_kontext_deblur.safetensors
|
30 |
+
Enter path to save patched LoRA .safetensors file: flux_kontext_deblur_nunchaku.safetensors
|
31 |
|
32 |
✅ Loaded 610 tensors from: Flux_kontext_deblur.safetensors
|
33 |
|
|
|
45 |
- lora_unet_final_layer_adaLN_modulation_1.lora_down.weight (shape: torch.Size([16, 3072]))
|
46 |
- lora_unet_final_layer_adaLN_modulation_1.lora_up.weight (shape: torch.Size([64, 16]))
|
47 |
|
48 |
+
✅ Patched file saved to: flux_kontext_deblur_nunchaku.safetensors
|
49 |
Total tensors now: 612
|
50 |
|
51 |
🔍 Verifying patched keys:
|