Update README.md
Browse files
README.md
CHANGED
|
@@ -40,6 +40,8 @@ For training I used the [Diffusion-Pipe](https://github.com/tdrussell/diffusion-
|
|
| 40 |
|
| 41 |
## Inference
|
| 42 |
|
|
|
|
|
|
|
| 43 |
For inference I recommend:
|
| 44 |
- [ComfyUI native nodes](https://blog.comfy.org/p/wan21-video-model-native-support)
|
| 45 |
- [Kijai's ComfyUI wrapper nodes](https://github.com/kijai/ComfyUI-WanVideoWrapper)
|
|
|
|
| 40 |
|
| 41 |
## Inference
|
| 42 |
|
| 43 |
+
It seems, the model likes long descriptive prompts. Look at the attached videos for prompt examples. Based on my tests if you use short prompts, the lora effect is weak.
|
| 44 |
+
|
| 45 |
For inference I recommend:
|
| 46 |
- [ComfyUI native nodes](https://blog.comfy.org/p/wan21-video-model-native-support)
|
| 47 |
- [Kijai's ComfyUI wrapper nodes](https://github.com/kijai/ComfyUI-WanVideoWrapper)
|