Save 4 hours! We auto-setup your workflow! Free!

Drop your workflow.json — we handle every dependency, custom node, and model. Just open the link and run.

Auto-Setup Workflow Json (Free) Now!
ComfyUI > Nodes > ComfyUI-LTX-FDG

ComfyUI Extension: ComfyUI-LTX-FDG

Repo Name

ComfyUI-LTX-FDG

Author
siraxe (Account age: 3323 days)
Nodes
View all nodes(1)
Latest Updated
2026-03-10
Github Stars
0.02K

How to Install ComfyUI-LTX-FDG

Install this extension via the ComfyUI Manager by searching for ComfyUI-LTX-FDG
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter ComfyUI-LTX-FDG in the search bar
After installation, click the Restart button to restart ComfyUI. Then, manually refresh your browser to clear the cache and access the updated list of nodes.

Visit ComfyUI Online for ready-to-use ComfyUI environment

  • Free trial available
  • 16GB VRAM to 80GB VRAM GPU machines
  • 400+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 200+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

ComfyUI-LTX-FDG Description

ComfyUI-LTX-FDG enhances ComfyUI and ComfyUI-LTXVideo by using Frequency-Decoupled Guidance (FDG), which applies distinct guidance scales to various frequency components, improving upon standard Classifier-Free Guidance.

ComfyUI-LTX-FDG Introduction

ComfyUI-LTX-FDG is an extension designed to enhance the capabilities of ComfyUI and ComfyUI-LTXVideo by implementing Frequency-Decoupled Guidance (FDG). This innovative approach allows AI artists to achieve high-fidelity sampling at low Classifier-Free Guidance (CFG) scales, which is particularly beneficial for maintaining the quality and diversity of generated content. By applying separate guidance scales to different frequency components, FDG helps in controlling the global structure and enhancing visual details without the drawbacks of high CFG values, such as oversaturation and loss of diversity.

How ComfyUI-LTX-FDG Works

The core principle of ComfyUI-LTX-FDG is based on the concept of frequency decomposition using Laplacian pyramids. Imagine breaking down an image into layers, each representing different levels of detail—from broad, sweeping shapes to fine, intricate textures. FDG applies different levels of guidance to these layers, allowing for precise control over the image's composition and detail. This method ensures that the global structure and fine details are both optimized, resulting in high-quality outputs even at lower CFG scales.

ComfyUI-LTX-FDG Features

  • Low-Frequency Guidance (w_low): This feature controls the overall structure, composition, and color of the output. By adjusting w_low, you can preserve the diversity of the generated content and prevent oversaturation. Lower values maintain diversity, while higher values improve alignment with the desired conditions.
  • High-Frequency Guidance (w_high): This feature enhances the visual fidelity, details, and sharpness of the output. It allows for higher values without the typical drawbacks associated with high CFG, such as color artifacts.
  • Frequency Levels: You can customize the number of frequency levels to balance between computational speed and output quality. More levels provide finer control but may slow down the process.

ComfyUI-LTX-FDG Models

ComfyUI-LTX-FDG supports different models tailored for specific use cases:

  • Video-Only (LTXV models): Ideal for generating video content with enhanced temporal stability and sharp details.
  • Video+Audio (LTXAV models): Supports both video and audio modalities, allowing for synchronized enhancements across both media types. This model uses separate FDGParameters nodes for video and audio, enabling tailored guidance for each.

What's New with ComfyUI-LTX-FDG

The latest updates to ComfyUI-LTX-FDG focus on improving user experience and output quality:

  • Enhanced Guidance Control: The ability to apply different guidance scales to frequency components has been refined, allowing for more precise control over the output.
  • Improved Performance: While FDG introduces some computational overhead, optimizations have been made to ensure that the extension runs efficiently without compromising quality.

Troubleshooting ComfyUI-LTX-FDG

Here are some common issues and solutions:

  • Oversaturation: If your videos appear oversaturated, try reducing w_low to a range between 1.0 and 1.5.

  • Blurry Details: Increase w_high to a range between 5.0 and 10.0 to enhance sharpness.

  • Flickering Frames: Reduce both w_low and w_high to improve temporal consistency.

  • FDG Not Working: Ensure that fdg_enabled is set to ON in the FDGParameters node and that all parameters are correctly connected to the MultimodalGuider.

  • Slow Sampling: To speed up the process, reduce frequency_levels to 1 or disable FDG by setting fdg_enabled to OFF.

Learn More about ComfyUI-LTX-FDG

For further exploration and support, consider the following resources:

  • Research Paper: Gain a deeper understanding of the theoretical foundation by reading the paper "Guidance in the Frequency Domain Enables High-Fidelity Sampling at Low CFG Scales" available here.
  • Community Forums: Engage with other AI artists and developers in community forums to share experiences, ask questions, and get support.
  • Documentation: Explore detailed documentation and tutorials to maximize your use of ComfyUI-LTX-FDG and its features. By leveraging these resources, you can enhance your creative projects and fully utilize the capabilities of ComfyUI-LTX-FDG.

ComfyUI-LTX-FDG Related Nodes

RunComfy
Copyright 2025 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals. RunComfy also provides AI Models, enabling artists to harness the latest AI tools to create incredible art.