ComfyUI > Nodes > Comfy_HunyuanImage3 > Hunyuan Latent Noise Shaping

ComfyUI Node: Hunyuan Latent Noise Shaping

Class Name

HunyuanLatentNoise

Category
Eric/HunyuanImage3/Latent
Author
EricRollei (Account age: 1544days)
Extension
Comfy_HunyuanImage3
Latest Updated
2026-02-21
Github Stars
0.05K

How to Install Comfy_HunyuanImage3

Install this extension via the ComfyUI Manager by searching for Comfy_HunyuanImage3
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter Comfy_HunyuanImage3 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

Hunyuan Latent Noise Shaping Description

HunyuanLatentNoise applies noise-shaping to alter image composition without changing prompts.

Hunyuan Latent Noise Shaping:

HunyuanLatentNoise is a versatile node designed to apply noise-shaping transformations to an existing Hunyuan latent. This node allows you to influence the composition of your generated images without altering the original prompt, providing a powerful tool for artistic expression and refinement. By utilizing various noise operations, such as low-pass and high-pass filters, spatial masking, and noise inversion, you can achieve a wide range of effects, from enhancing macro compositions to introducing intricate details. The node's primary goal is to offer you the flexibility to manipulate the latent space creatively, enabling you to explore different artistic styles and compositions effortlessly.

Hunyuan Latent Noise Shaping Input Parameters:

latent

The latent parameter represents the input latent space that you wish to transform. It is a crucial component as it serves as the foundation upon which all noise operations are applied. This parameter is expected to be of type HUNYUAN_LATENT, ensuring compatibility with the node's processing capabilities.

operation

The operation parameter specifies the type of noise-shaping transformation to apply to the latent. It offers a selection of operations, including "low_pass_filter," "high_pass_filter," "blend_seeds," "spatial_mask_left_right," "spatial_mask_top_bottom," "amplify_center," and "invert." Each operation has a unique impact on the latent, allowing you to tailor the output to your artistic vision. The default operation is "low_pass_filter."

strength

The strength parameter controls the intensity of the selected noise operation. It is a floating-point value ranging from 0.0 to 1.0, with a default value of 0.5. Adjusting the strength allows you to fine-tune the effect of the operation, providing subtle or pronounced transformations based on your preference.

seed

The seed parameter is an integer value used to initialize the random number generator for noise operations that require randomness, such as "blend_seeds" and spatial masking. It ensures reproducibility of results by allowing you to generate consistent noise patterns across different runs. The default seed value is 42, with a valid range from 0 to 2^63

  • 1.

Hunyuan Latent Noise Shaping Output Parameters:

latent

The latent output parameter provides the transformed latent space after applying the specified noise operation. This output retains the original dimensions of the input latent while incorporating the desired noise effects. It serves as the basis for further processing or image generation, allowing you to visualize the impact of the noise transformations on your creative work.

Hunyuan Latent Noise Shaping Usage Tips:

  • Experiment with different operation settings to discover unique artistic effects and compositions that align with your creative vision.
  • Use the strength parameter to control the subtlety or intensity of the noise transformations, enabling you to achieve the desired balance between detail and composition.
  • Utilize the seed parameter to ensure consistency in your results, especially when working on iterative projects or when you need to reproduce specific noise patterns.

Hunyuan Latent Noise Shaping Common Errors and Solutions:

Noise shape mismatch

  • Explanation: This error occurs when the shape of the noise tensor does not match the shape of the latent tensor, leading to an inability to apply the noise operation correctly.
  • Solution: Ensure that the noise tensor and latent tensor have compatible shapes. If necessary, adjust the dimensions of the noise tensor or regenerate it to match the latent tensor's shape.

Invalid operation selection

  • Explanation: This error arises when an unsupported or misspelled operation is specified in the operation parameter.
  • Solution: Verify that the operation parameter is set to one of the supported operations: "low_pass_filter," "high_pass_filter," "blend_seeds," "spatial_mask_left_right," "spatial_mask_top_bottom," "amplify_center," or "invert." Correct any typos or unsupported values.

Hunyuan Latent Noise Shaping Related Nodes

Go back to the extension to check out more related nodes.
Comfy_HunyuanImage3
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.

Hunyuan Latent Noise Shaping