Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,24 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
base_model: Wan-AI/Wan2.1-T2V-1.3B
|
| 4 |
+
tags:
|
| 5 |
+
- wan
|
| 6 |
+
- video
|
| 7 |
+
- text-to-video
|
| 8 |
+
- diffusion-pipe
|
| 9 |
+
- lora
|
| 10 |
+
- template:sd-lora
|
| 11 |
+
- standard
|
| 12 |
+
library_name: diffusers
|
| 13 |
+
pipeline_tag: text-to-video
|
| 14 |
+
---
|
| 15 |
+
## CausVid Accelerator Misubi-format Conversion
|
| 16 |
+
## Text/Image to Video Low Rank Adapter (LoRA)
|
| 17 |
+
## For Wan2.1 1.3B family models (Base, Skyreels V2, FunINP1.3B, etc...)
|
| 18 |
+
|
| 19 |
+
Based on [Kijal's original ComfyUI-oriented LoRA extraction](https://huggingface.co/Kijai/WanVideo_comfy/blob/main/Wan21_CausVid_bidirect2_T2V_1_3B_lora_rank32.safetensors) of the CausVid accelerated mod of Wan2.1 1.3B. <br>
|
| 20 |
+
This conversion **might** work better than Kijal's source for those using CausVid outside of a ComfyUI environment. <br>
|
| 21 |
+
**Try this version if:** Kijal's original LoRA does not work **at all** for you and crashes/fails to infer. <br>
|
| 22 |
+
*For example:* This version **works in DrawThings**, whereas Kijal's original does not (as of the time I write this) & simply crashes. <br>
|
| 23 |
+
This might (or might not) extend to other given non-Comfy Wan-inference environments/backends/frameworks. <br>
|
| 24 |
+
Please share any other confirmed use cases in the "Community" tab! <br>
|