ComfyUI-LTXVideo

ComfyUI-LTXVideo is a plugin for ComfyUI, a visual tool people use to build AI-powered image and video generation workflows by connecting nodes on a canvas. This plugin adds support for LTX-2, a video generation model made by Lightricks, the company behind video editing apps. With it installed, you can generate videos from text descriptions or from a starting image.

The core idea is text-to-video and image-to-video generation. You write a description of what you want to see, or provide a still image as the starting frame, and the model produces a short video clip. There are two processing modes: a full-quality mode that takes longer and a distilled mode that runs faster at a slight quality trade-off. Both are included as ready-made workflow files you can load directly into ComfyUI.

Beyond basic video generation, the plugin includes several specialized add-ons called IC-LoRAs. One lets you control the generated video using depth maps or edge outlines from a reference image, so motion follows a specific structure. Another generates high-dynamic-range video suitable for professional color grading and export as EXR image sequences. A third, called Lipdub, can reanimate a person's lips in a video to match new dialogue, either in the original language or translated into another.

Installing it requires ComfyUI to already be set up, a compatible GPU with at least 32 gigabytes of video memory, and around 100 gigabytes of free disk space for the model files. The installation itself goes through ComfyUI's built-in Manager tool: search for LTXVideo, click install, and restart. The required model files download automatically on first use.

The LTX-2 model is built directly into ComfyUI's core, so this repository specifically provides the extra nodes and pre-built workflow files that make the advanced features accessible. It is aimed at people who already use ComfyUI and want to add video generation to their existing setup.