Deflate Open-Source Model - Add deflating effects to objects in images and easily generate dynamic videos

Deflate

Developed by Remade-AI

A LoRA trained on the Wan2.1 14B I2V 480p model that adds deflation effects to any object in an image to generate dynamic videos.

Text-to-Video EnglishOpen Source License:Apache-2.0 #Object Deflation Animation #Image to Video #LoRA Effects

Downloads 1,664

Release Time : 3/11/2025

Model Overview

This LoRA model can transform static images into dynamic videos featuring object deflation, applicable to various subjects including animals, vehicles, and characters.

Model Features

Cross-Object Type Consistency

Capable of generating consistent deflation effects for various types of objects (e.g., animals, characters, vehicles).

Simple Prompt Structure

Uses straightforward prompt templates for effect generation, making it easy to adjust and customize.

High-Quality Video Generation

Trained on the Wan2.1 14B I2V 480p base model, generating dynamic videos at 480p resolution.

Model Capabilities

Image to Video Generation

Object Morphing Effects

Dynamic Video Synthesis

Use Cases

Creative Video Production

Character Deflation Effect

Generates a video effect where a character gradually deflates, shrinks, and deforms.

example_videos/man_deflate.mp4

Object Deflation Effect

Generates deflation and deformation effects for various objects (e.g., lamps, balloons).

example_videos/lamp_deflate.mp4, example_videos/balloon_deflate.mp4

Animal Deflation Effect

Generates a video effect where an animal (e.g., a cat) deflates, shrinks, and deforms.

example_videos/cat_deflate.mp4

🚀 Deflate Effect LoRA for Wan2.1 14B I2V 480p

This LoRA is trained on the Wan2.1 14B I2V 480p model, enabling you to deflate any object in an image. It can be applied to a wide range of objects, including animals, vehicles, and people.

🚀 Quick Start

This LoRA allows you to transform images into deflated videos. You can join the Discord community to generate videos for free and also request more Wan2.1 LoRAs.

✨ Features

Transform any image into a video of it being deflated.
Trained on the Wan2.1 14B 480p I2V base model.
Consistent results across different object types.
Simple prompt structure that's easy to adapt.

📦 Installation

📥 Download Links:

deflate_20_epochs.safetensors - LoRA Model File
wan_img2vid_lora_workflow.json - Wan I2V with LoRA Workflow for ComfyUI

📚 Documentation

Recommended Settings

Property	Details
LoRA Strength	1.0
Embedded Guidance Scale	6.0
Flow Shift	5.0

Trigger Words

The key trigger phrase is: d3d1at3 deflate it

Prompt Template

For best results, use this prompt structure:

The video opens with a [object]. As the video progresses, the [object] begins to d3d1at3 deflate it, gradually shrinking and losing shape, eventually flattening completely into a lifeless, deflated mass on the ground.

Simply replace [object] with whatever you want to see deflated!

ComfyUI Workflow

This LoRA works with a modified version of Kijai's Wan Video Wrapper workflow. The main modification is adding a Wan LoRA node connected to the base model.

Workflow Screenshot

See the Downloads section above for the modified workflow.

Model Information

The model weights are available in Safetensors format. See the Downloads section above.

Training Details

Property	Details
Base Model	Wan2.1 14B I2V 480p
Training Data	Trained on 1.5 minutes of video comprised of 20 short clips (each clip captioned separately) of things being deflated
Epochs	20

Additional Information

Training was done using Diffusion Pipe for Training

Acknowledgments

Special thanks to Kijai for the ComfyUI Wan Video Wrapper and tdrussell for the training scripts!

📄 License

This project is licensed under the Apache-2.0 license.

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご