Rachel Luxemburg
8b2d38c2ec
Moving the installation instructions from the readme file to a separate file for better discoverability.
3.2 KiB
Getting Started
Although sample nodes and a workflow can be found within ComfyUI, for best results we recommend the below process:
Installation
- Install ComfyUI
- Clone this repository to
custom-nodesfolder in your ComfyUI installation directory. - Install the required packages:
cd custom_nodes/ComfyUI-LTXVideo && pip install -r requirements.txt
For portable ComfyUI installations, run
.\python_embeded\python.exe -m pip install -r .\ComfyUI\custom_nodes\ComfyUI-LTXVideo\requirements.txt
Models
- Download ltx-video-2b-v0.9.1.safetensors from Hugging Face and place it under
models/checkpoints. - Install a t5 text encoder. We recommend google_t5-v1_1-xxl_encoderonly. You can install it using ComfyUI Model Manager.
Example Workflows
Select a workflow from the examples and download it.
Note that to run the example workflows below, you need to have some additional custom nodes, like ComfyUI-VideoHelperSuite and others, installed. You can add them by pressing "Install Missing Custom Nodes" button in ComfyUI Manager.
Long Video Generation
🧩 Image to Video Long Video: Long video generation with support for multiple prompts along the video duration.
🧩 Video to Video Long Video: Long video-to-video generation. Given a guiding video—such as depth, pose, or edges—the flow generates a new video.
Video Upscaling
🧩 Video Upscaling: Upscales and adds fine details to any given video, increasing its spatial resolution by 2×.
Easy to use multi scale generation workflows
🧩 Image to video mixed: mixed flow with full and distilled model for best quality and speed trade-off.
13B model
🧩 Image to video
🧩 Image to video with keyframes
🧩 Image to video with duration extension
🧩 Image to video 8b quantized
13B distilled model
🧩 Image to video
🧩 Image to video with keyframes
🧩 Image to video with duration extension
🧩 Image to video 8b quantized
