starsfriday LoRA for LTX-2

Overview

This LoRA is trained on the LTX-2 model.

Features

• Make any start and end frames of the video transition smoothly
• Trained on the LTX-2 base model
• Consistent results across different object types
• Simple prompt structure that's easy to adapt

Model File and Inference Workflow

📥 Model File:

Recommended Settings

• LoRA Strength: 1.0
• Embedded Guidance Scale: 1.0
• Classifier Free Guidance: 4.0

Trigger Words

The key trigger phrase is: huanzhuang

Prompt Template

For best results, use this prompt structure:

The video showcases a highly dynamic horizontal displacement costume display. The character's arms are raised above his head and crossed. As his hands fall, the new look swiftly sweeps from one side of the frame to the other. Through this rapid rotation and encireption, it gradually covers the original costume from top to bottom along the body. huanzhuang skillfully completed the gradient and replacement of the clothing in the high-speed flowing afterimage,huanzhuang