ComfyUI  >  Nodes  >  smZNodes

ComfyUI Extension: smZNodes

Repo Name


shiimizu (Account age: 1774 days)
View all nodes (2)
Latest Updated
Github Stars

How to Install smZNodes

Install this extension via the ComfyUI Manager by searching for  smZNodes
  • 1. Click the Manager button in the main menu
  • 2. Select Custom Nodes Manager button
  • 3. Enter smZNodes 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
  • High-speed GPU machines
  • 200+ preloaded models/nodes
  • Freedom to upload custom models/nodes
  • 50+ ready-to-run workflows
  • 100% private workspace with up to 200GB storage
  • Dedicated Support

Run ComfyUI Online

smZNodes Description

smZNodes enables ComfyUI to generate identical embeddings as stable-diffusion-webui using the CLIP Text Encode++ node.

smZNodes Introduction

ComfyUI_smZNodes is an extension for that provides a selection of custom nodes designed to enhance your AI art generation experience. This extension allows you to generate identical embeddings from within ComfyUI, enabling you to reproduce the same images across both platforms. Whether you are working with simple or complex prompts, ComfyUI_smZNodes helps you achieve consistent and high-quality results.

How smZNodes Works

ComfyUI_smZNodes works by providing custom nodes that integrate with ComfyUI to process text prompts and generate embeddings. These embeddings are then used to create images. The extension includes nodes like CLIP Text Encode++ and Settings, which allow you to fine-tune the text encoding and sampling process. By using these nodes, you can ensure that the images generated in ComfyUI match those created in stable-diffusion-webui, even when using complex prompts with attention, emphasis, and weighting.

smZNodes Features

CLIP Text Encode++

The CLIP Text Encode++ node is a powerful tool that generates identical embeddings from stable-diffusion-webui for use in ComfyUI. This ensures that the images you create in ComfyUI are consistent with those generated in stable-diffusion-webui.

Key Features:

  • Prompt Editing: Modify your prompts to include alternating words and specific keywords like AND and BREAK.
  • Weight Normalization: Normalize the weights of your prompts to ensure balanced emphasis.
  • Optional Embedding Identifier: Use an optional embedding: identifier for more control.


  • Parser: Choose from different parsers to transform prompts into tokens and embeddings.
  • Mean Normalization: Decide whether to take the mean of your prompt weights.
  • Multi-Conditioning: Use multiple prompts at once, allowing for complex compositions.

Settings Node

The Settings node is a dynamic tool that allows you to fine-tune results during sampling or tokenization. It functions similarly to the Reroute node and can be customized even after being connected.

Key Features:

  • Dynamic Inputs: Replace inputs with another type even after connection.
  • Local Settings Application: Apply settings locally based on links, similar to model patches.
  • Override Capabilities: Override settings by using another Settings node further down the workflow.

smZNodes Models

ComfyUI_smZNodes does not include different models but provides various parsers and settings to customize how text prompts are processed and encoded. These options allow you to tailor the behavior of the extension to match your specific needs.

Available Parsers:

  • comfy: Default parser for ComfyUI.
  • comfy++: Uses ComfyUI's parser but encodes tokens like stable-diffusion-webui.
  • A1111: Default parser used in stable-diffusion-webui.
  • full: Similar to A1111 but strips whitespaces, newlines, and special characters.
  • compel: Uses the library.
  • fixed attention: Leaves the prompt untampered.

Troubleshooting smZNodes

Common Issues and Solutions

  1. Images Not Matching Between UIs:
  • Ensure you are using the same seed, sampler settings, RNG (CPU or GPU), and clip skip settings.
  • For DDIM sampler, use the ddim_uniform scheduler.
  1. Settings Node Issues:
  • If the Settings node changes, you may need to recreate it to prevent issues.
  • Use another Settings node to override previous settings if necessary.


  • How does this differ from ComfyUI_ADV_CLIP_emb?
  • While both normalize weights similarly, the tokenization and encoding pipeline from stable-diffusion-webui in ComfyUI_smZNodes produces different results.
  • Where can I learn more about how ComfyUI interprets weights?
  • ComfyUI Guides (

Learn More about smZNodes

For additional resources, tutorials, and community support, you can explore the following:

  • ComfyUI Community Forum ( These resources provide valuable information and support to help you make the most out of ComfyUI_smZNodes and enhance your AI art generation experience.

smZNodes Related Nodes


© Copyright 2024 RunComfy. All Rights Reserved.

RunComfy is the premier ComfyUI platform, offering ComfyUI online environment and services, along with ComfyUI workflows featuring stunning visuals.