Quick Start

Let’s get started generating video!

Choose Your Workflow

LTX-2.3 offers multiple ways to generate videos locally. We recommend starting with ComfyUI for the best balance of power and ease of use. If you prefer a standalone app with a built-in video editor, choose LTX Desktop.

ComfyUI Quick Start

ComfyUI provides an intuitive node-based interface. If you have not previously installed ComfyUI, download it here: https://www.comfy.org/download

Installation

Recommended: ComfyUI Manager

1. Launch ComfyUI and launch ComfyUI Manager
2. Search for “LTXVideo”
3. Click Install on the ComfyUI-LTXVideo node pack
4. Restart ComfyUI

Alternative: Manual Installation

$
# Clone the repository
$
git clone https://github.com/Lightricks/ComfyUI-LTXVideo.git
$
cd ComfyUI-LTXVideo
$
$
# Set up the environment
$
uv sync --frozen
$
source .venv/bin/activate

For portable/embedded installations:

$
.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\ComfyUI-LTXVideo\requirements.txt

Your First Video

1. Load a workflow: In ComfyUI, click Load and select an example workflow from the LTXVideo nodes
2. Set your prompt: Find the text encode node and enter your description
3. Queue prompt: Click Queue Prompt to generate

Video generation time will depend on your GPU and VRAM.

LTX Desktop

LTX Desktop is an open-source desktop app for generating videos with LTX models locally on supported Windows/Linux NVIDIA GPUs, with an API mode for unsupported hardware and macOS.

Download the latest release from https://github.com/Lightricks/LTX-Desktop/releases

Requirements

You’ll also need at least 16 GB system RAM (32 GB recommended) and 160 GB free disk space for model weights, the Python environment, and outputs.

Installation

1. Download the latest installer from GitHub Releases
2. Install and launch LTX Desktop
3. Complete first-run setup — you’ll be prompted to accept model license terms
4. Configure text encoding (choose one):
   • LTX API key (recommended) — Cloud-based text encoding is free and speeds up inference.
   • Local Text Encoder — Enable in Settings for fully offline operation. Requires an additional model download.

Your First Video

1. Enter a text prompt describing the video you want to generate
2. Click Generate

Generation time depends on your GPU, file resolution, and duration.

What’s Next?

Now that you have LTX-2.3 running, explore these guides to get the most out of the model:

Essential Guides

Getting Help