LTX 2.3 Depth-Guided Video-to-Video Workflow

Workflow Overview

This workflow uses an input motion video to guide the first-frame reference image, allowing the generated video to reproduce the same character movements and camera motion.

Model Setup

First, download the base LTX 2.3 models. The model requirements are the same as those for other LTX workflows. Follow the instructions at the link below:

https://dfcine.com/2549.html

In addition, download the following control LoRA:

ltx-2.3-22b-ic-lora-union-control-ref0.5.safetensors

Place it in:

ComfyUI\models\loras

Download link:

https://huggingface.co/Lightricks/LTX-2.3-22b-IC-LoRA-Union-Control/tree/main