Create README.md
Browse files
README.md
ADDED
|
@@ -0,0 +1,73 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
# Squish Effect LoRA for Wan2.1 14B I2V 480p
|
| 2 |
+
|
| 3 |
+

|
| 4 |
+
|
| 5 |
+
## Overview
|
| 6 |
+
|
| 7 |
+
This LoRA is trained on the Wan2.1 14B I2V 480p model and allows you to squish any object in an image. The effect works on a wide variety of objects, from animals to vehicles to people!
|
| 8 |
+
|
| 9 |
+
## Features
|
| 10 |
+
|
| 11 |
+
- Transform any image into a video of it being squished
|
| 12 |
+
- Trained on the Wan2.1 14B 480p I2V base model
|
| 13 |
+
- Consistent results across different object types
|
| 14 |
+
- Simple prompt structure that's easy to adapt
|
| 15 |
+
|
| 16 |
+
## Examples
|
| 17 |
+
|
| 18 |
+
Here are some examples of the squish effect in action:
|
| 19 |
+
|
| 20 |
+
| Object | Video |
|
| 21 |
+
|--------|-------|
|
| 22 |
+
| Clay Dog | [View Example](https://placeholder-link.com) |
|
| 23 |
+
| Tank | [View Example](https://placeholder-link.com) |
|
| 24 |
+
| Hot Air Balloon | [View Example](https://placeholder-link.com) |
|
| 25 |
+
| Rodent | [View Example](https://placeholder-link.com) |
|
| 26 |
+
| Woman | [View Example](https://placeholder-link.com) |
|
| 27 |
+
|
| 28 |
+
## Usage
|
| 29 |
+
|
| 30 |
+
### Recommended Settings
|
| 31 |
+
|
| 32 |
+
- **LoRA Strength:** 1.0
|
| 33 |
+
- **Embedded Guidance Scale:** 6.0
|
| 34 |
+
- **Flow Shift:** 5.0
|
| 35 |
+
|
| 36 |
+
### Trigger Words
|
| 37 |
+
|
| 38 |
+
The key trigger phrase is: `sq41sh squish effect`
|
| 39 |
+
|
| 40 |
+
### Prompt Template
|
| 41 |
+
|
| 42 |
+
For best results, use this prompt structure:
|
| 43 |
+
|
| 44 |
+
_In the video, a miniature [object] is presented. The [object] is held in a person's hands. The person then presses on the [object], causing a sq41sh squish effect. The person keeps pressing down on the [object], further showing the sq41sh squish effect._
|
| 45 |
+
|
| 46 |
+
Simply replace `[object]` with whatever you want to see squished!
|
| 47 |
+
|
| 48 |
+
## ComfyUI Workflow
|
| 49 |
+
|
| 50 |
+
This LoRA works with a modified version of [Kijai's Wan Video Wrapper workflow](https://github.com/kijai/ComfyUI-WanVideoWrapper/blob/main/example_workflows/wanvideo_480p_I2V_example_02.json). The main modification is adding a Wan LoRA node connected to the base model.
|
| 51 |
+
|
| 52 |
+

|
| 53 |
+
|
| 54 |
+
## Download
|
| 55 |
+
|
| 56 |
+
Weights for this model are available in Safetensors format.
|
| 57 |
+
|
| 58 |
+
[Download](https://huggingface.co/yourusername/squish-effect-lora/tree/main) them in the Files & versions tab.
|
| 59 |
+
|
| 60 |
+
## Community
|
| 61 |
+
|
| 62 |
+
- **Discord:** [Join our community](https://discord.com/invite/7tsKMCbNFC) to generate videos with this LoRA for free
|
| 63 |
+
- **Request LoRAs:** We're training and open-sourcing Wan2.1 LoRAs for free - join our Discord to make requests!
|
| 64 |
+
|
| 65 |
+
## Training Details
|
| 66 |
+
|
| 67 |
+
- **Base Model:** Wan2.1 14B I2V 480p
|
| 68 |
+
- **Training Data:** 1.5 minutes of video (20 short clips of things being squished)
|
| 69 |
+
- **Epochs:** 18
|
| 70 |
+
|
| 71 |
+
## Acknowledgments
|
| 72 |
+
|
| 73 |
+
Special thanks to Kijai for the ComfyUI Wan Video Wrapper that makes using this LoRA possible!
|