ComfyUI > Nodes > ComfyUI-LG_SamplingUtils > 🎈LG Noise Injection (Latent)

ComfyUI Node: 🎈LG Noise Injection (Latent)

Class Name

LGNoiseInjectionLatent

Category
advanced/model
Author
LAOGOU-666 (Account age: 656days)
Extension
ComfyUI-LG_SamplingUtils
Latest Updated
2025-12-24
Github Stars
0.14K

How to Install ComfyUI-LG_SamplingUtils

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

🎈LG Noise Injection (Latent) Description

Enhances AI images by injecting desired features into latent space for detailed realism.

🎈LG Noise Injection (Latent):

LGNoiseInjectionLatent is a powerful node designed to enhance your AI-generated images by injecting specific features directly into the latent space during the CFG (Classifier-Free Guidance) process. This node allows you to input a reference latent, which contains the desired characteristics you wish to incorporate into your output. It is particularly useful for adding intricate surface details such as water droplets, sweat, textures, material feel, gloss, and reflection effects. By leveraging the latent features of a reference image, this node enables you to achieve a more nuanced and detailed output, enhancing the visual richness and realism of your AI-generated art.

🎈LG Noise Injection (Latent) Input Parameters:

model

This parameter specifies the model to be used for the noise injection process. It is crucial as it determines the underlying architecture and capabilities that will be applied to the latent space. The model acts as the foundation upon which the reference latent features are injected.

reference_latent

The reference_latent parameter is a critical input that contains the latent features you wish to inject into the CFG process. This latent should embody the characteristics or details you want to see in the final output, such as specific textures or reflections. It serves as the blueprint for the feature injection, guiding the transformation of the generated image.

strength

The strength parameter controls the intensity of the feature injection, with a default value of 0.15. It ranges from 0.0 to 1.0, where lower values (0.1-0.2) result in subtle enhancements, and higher values (0.2-0.4) produce more pronounced effects. Adjusting this parameter allows you to fine-tune the balance between the original image and the injected features, ensuring the desired level of detail and impact.

🎈LG Noise Injection (Latent) Output Parameters:

injected_latent

The injected_latent output represents the modified latent space after the reference features have been injected. This output is crucial as it forms the basis for generating the final image, now enriched with the desired characteristics from the reference latent. It reflects the successful integration of new features, providing a more detailed and visually appealing result.

🎈LG Noise Injection (Latent) Usage Tips:

  • To achieve subtle enhancements, start with a lower strength value (around 0.1) and gradually increase it to see how it affects the output. This approach helps in maintaining control over the feature injection process.
  • Use a reference latent that closely matches the characteristics you want to inject. The more aligned the reference is with your desired outcome, the more effective the feature injection will be.

🎈LG Noise Injection (Latent) Common Errors and Solutions:

"Shape mismatch between reference and CFG result"

  • Explanation: This error occurs when the dimensions of the reference latent do not match those of the CFG result, preventing proper interpolation.
  • Solution: Ensure that the reference latent is resized to match the CFG result dimensions using appropriate interpolation methods as described in the node's functionality.

"Invalid strength value"

  • Explanation: This error arises when the strength parameter is set outside the allowed range of 0.0 to 1.0.
  • Solution: Adjust the strength value to fall within the specified range, ensuring it is neither negative nor exceeds 1.0.

🎈LG Noise Injection (Latent) Related Nodes

Go back to the extension to check out more related nodes.
ComfyUI-LG_SamplingUtils
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.