ComfyUI-DD-Nodes

中文 | English

🚀 Project Overview

ComfyUI-DD-Nodes is a professional ComfyUI custom node collection focused on improving AI generation workflow efficiency and user experience. This project adopts a full-process AI development approach, providing a fully localized UI and rich functional modules, aiming to deliver the best ComfyUI extension experience for Chinese users.

✨ Core Features

• 🎯 Professional Toolkit: Covers prompt management, model optimization, image processing, UI enhancement, and more
• 🌏 Fully Localized: Supports ComfyUI native internationalization (i18n) with seamless Chinese-English switching
• ⚡ Performance Optimization: Multiple optimizations including intelligent model loading and animation performance tuning
• 🎨 UI Enhancement: Theme system, connection animations, layout tools, and other visual enhancement features
• 🔧 Easy to Use: Clean and intuitive UI, reducing the learning curve
• 📈 Actively Maintained: Continuous development with fast response to user needs

🎯 Extension Functions Overview

Function Name	Function Description	Image Description
DD Prompt Manager	An intelligent prompt embedding and management system. It provides powerful prompt management for all CLIP text encoder nodes, including search, tag categorization, import/export, real-time sync, and more — significantly improving workflow creation efficiency
DD Interface Layout	An intelligent node layout tool. Supports random or specified node title colors, node alignment, size synchronization, and more. Built-in theme system (Chaos Mode / Order Mode), each theme has two styles. Click the coin animation to switch. Layout supports real-time preview and animation
DD Node Connection Animation	Provides multiple cool animation effects for ComfyUI node connections, including "Flow", "Wave", "Rhythm", "Pulse", etc. Supports 4 rendering styles (straight, right-angle, curve, circuit board). Parameters are customizable (width, speed, effects, etc.). Provides "Show All" and "Hover Node" modes; hover mode uses static + animated overlay rendering for smooth visuals

📋 Node Feature Preview

Node Name	Function Description	Image Description
DD Image Outline	Intelligent image outlining tool. Supports outlining for transparent and normal images. Provides outer outline, inner outline, and centered outline modes, with customizable color, size, opacity, and mask support
DD Qwen-MT Translator	Powerful multilingual translation node based on Alibaba Cloud Qwen-MT. Supports translation between 92 languages and provides General, Terminology, and Domain translation modes
DD Model Optimized Loading	High-performance model loading optimizer supporting intelligent loading and multiple optimization modes. Built-in Smart Mode automatically selects the best loading strategy based on model size and hardware configuration. Supports all models loaded via UNET nodes
DD Image To Video Frames	Efficient image-to-video-frames converter supporting batch processing and multiple output formats, providing convenience for video generation workflows
DD Advanced Fusion	Powerful image and video fusion processor supporting multiple fusion algorithms and parameter tuning to achieve professional-grade compositing effects
DD Dimension Calculator	Minimal image dimension calculator providing precise size calculation and ratio adjustment to ensure outputs match expected specs
DD Simple Latent	Simplified latent generator that provides a fast and convenient way to create latent space, optimizing workflow node connections
DD Image Uniform Size	Multi-functional image/video uniform-sizing processor with batch processing and smart scaling, ensuring consistent output dimensions
DD Mask Uniform Size	Professional mask uniform-sizing tool that pairs with image processing to ensure mask dimensions match the target image
DD Image Size Limiter	Smart image size limiter ensuring images stay within configured max/min sizes to prevent memory overflow and performance issues
DD Switcher Series	A set of switcher nodes including Conditional Switcher, Latent Switcher, Model Switcher, etc., simplifying workflows and improving flexibility
DD Video First/Last Frame Output	Video frame extraction tool that precisely outputs the first and last frames of a video for video workflows
DD Image Splitter	Intelligent image splitter supporting custom ratio splitting. Choose left-right or top-bottom splitting, supports unequal ratios (e.g., 2:1:3), and outputs the specified split part
DD Aspect Ratio Selector	Aspect ratio selection tool that provides recommended resolutions for different models (e.g., Qwen-image, Wan2.2). Supports landscape/portrait/square categories and automatically provides the most suitable sizes
DD TXT File Merger	Powerful text merger that recursively scans all TXT files in a folder (and subfolders) and merges them into a single text. Automatically adds file-path markers and supports multiple encodings

🔧 Installation Methods

Method 1: Manual Installation

1. Copy files to ComfyUI's custom_nodes directory
2. Install dependencies: pip install -r requirements.txt
3. Restart ComfyUI

Method 2: Manager Installation (Recommended)

1. Use ComfyUI Manager or launcher for git installation
2. Or search "ComfyUI-DD-Nodes" directly in Manager and install

🌐 Language Switching

This extension now supports ComfyUI's native internationalization (i18n) system for seamless Chinese-English switching:

1. Switch Language: Select language options in the settings menu in the upper right corner of ComfyUI interface
2. Select Language: Supports Chinese (zh) and English (en)
3. Function Scope: Currently supports node translation, frontend function translation is under development

📈 Version History

Latest Version

• v3.1.3 (2025-01-24)
• Connection animation speed changed from 3 levels to 6 levels

• Text in the settings panel now switches with language (currently only Chinese and English)

• v3.1.2 (2025-12-29)

• Added “Show connections for selected nodes” to the connection animation display system

• v3.1.1 (2025-12-10)

• Added compatibility support for Vue Nodes 2.0 nodes

• v3.1.0 (2025-09-21)

• Added DD TXT File Merger node, supporting recursive scan of TXT files and merging into one text

• v3.0.0 (2025-09-05)

• Added DD Image Splitter node, supporting custom ratio splitting and output of specified parts

• v2.9.1 (2025-09-04)

• Removed DD Sampler node (it no longer fits the latest ComfyUI)

• v2.9.0 (2025-09-01)

• Added Aspect Ratio Selector node

• v2.8.3 (2025-08-23)

• Prompt Manager added embedding support for the TextEncodeQwenImageEdit node

• v2.8.2 (2025-08-01)

• Removed Circuit Board 1 (it is effectively the same as right-angle). If connections disappear after update, just re-select the style

• Fully synced path calculation for straight/curve/right-angle styles with ComfyUI official implementation

• v2.8.1 (2025-07-31)

• Added default rendering style selection for static lines, supporting official-static+dynamic and independent-static+dynamic modes

• v2.8.0 (2025-07-30)

• New: Qwen-MT Translation:
  • Integrated Alibaba Cloud Qwen-MT, supports translation between 92 languages
  • Supports General / Terminology / Domain translation modes
  • Provides a convenient API key configuration UI, supporting one-click configuration and management
• Major connection animation system upgrade:
  • Optimized visuals in hover mode, using fully independent static-style rendering
  • In hover mode, inactive connections now render as static; active connections overlay animation effects

• Circuit Board 2 performance optimization:

  • Spatial partitioning optimization greatly improves path calculation performance (50–70% for small graphs, 80–95% for large graphs)
  • Dynamic grid size adjustment to optimize performance based on node count

• v2.7.0 (2025-07-26)

• 🎨 Added DD Image Outline node, supporting intelligent outlining for transparent and normal images
• 🎯 Supports outer/inner/center outline modes for different use cases

• 🎮 Supports custom outline size and opacity, fully compatible with masks

• v2.6.0 (2025-07-06)

• 🔧 Attempted to fix multi-OS compatibility issues for the Interface Layout panel

• ➕ Added Video First/Last Frame Output node

• v2.5.1 (2025-07-02)

• 🎯 Prompt Manager panel perfectly embeds most official and third-party CLIP nodes
• 🔗 Perfectly compatible with ComfyUI-Easy-Use and ComfyUI-Fast-Use

• 🐛 Fixed some known issues

• v2.5.0 (2025-06-23)

• 🎨 Full UI/interaction refactor for Prompt Manager
• 🏷️ Tag colors are freely editable, with batch edit support
• 🔍 Search upgrade: tag + text dual search
• 💾 Real-time sync to local file, no more browser cache

• 📤 Export includes complete data (prompts + tags)

• v2.4.0 (2025-06-20)

• 🔄 Prompt Manager new features:

  • Smart backup: all prompt operations are saved to extensions\Prompt_Manager\prompts.json
  • Added tags: up to 4 tags per prompt for categorization and quick lookup

• v2.3.0 (2025-06-18)

• 🎉 New: Prompt Manager:
  • Added smart prompt embedding for all CLIP text encoder nodes
  • Supports CRUD for prompts with a modern management UI
  • Supports import/export of prompts in JSON
  • Supports searching by prompt name and content

• ⚡ Online Animation performance & feature optimization:

  • Added visual options: "Show All" and "Hover Node"
  • Improved smart FPS control: dynamically adjusts redraw frequency based on connection count (hover mode is close to native FPS)

• v2.2.1 (2025-05-27)

• Improved UI, added theme options, and fixed bugs

• v2.2.0 (2025-05-14)

• Interface Layout Alpha 2.0 released with improved UI and basic tools

• Can now sync width/height/size of selected nodes and align all selected nodes

• v2.1.0 (2025-05-13)

• Added support for ComfyUI native i18n
• Seamless Chinese-English switching
• All node names, descriptions, parameters, and outputs support bilingual display

Click to view more version history

• v2.0.2 (2025-05-12)

• Further improved animation performance and effects, fixed some path bugs

• v2.0.1 (2025-05-08)

• Fixed compatibility issues between connection animation and cg-use-everywhere

• Fixed a serious bug in the Interface Layout feature

• v2.0.0 (2025-05-08)

• Modularized connection animation effects for easier future maintenance
• Optimized animation performance and algorithms, added animation sampling strength options

• [New Feature!] Implemented Alpha version of Interface Layout

• v1.9.1 (2025-05-05~06)

• Permanently removed node alignment functionality and all its code

• [Project changed to MIT license, fully open-sourced]

• v1.9.0 (2025-05-04)

• Added "Circuit Board 1" and "Circuit Board 2" frontend connection styles
• Circuit Board 1: classic L-shaped / folded paths, compatible with any node layout, animations and effects fully supported

• Circuit Board 2: supports recursive obstacle avoidance, prefers 90/45-degree connections, and intelligently routes around obstacle nodes

• v1.8.0 (2025-05-03)

• Added a complete set of frontend connection animations

• v1.6.0 (2025-04-21)

• Added Mask Uniform Size node
• Added Image Size Limiter node

• Added Switcher Series nodes (Conditional Switcher, Latent Switcher, Model Switcher)

• v1.5.0 (2025-03-05)

• Added Image Uniform Size node

• v1.3.0 (2025-03-01)

• Added a Sampling Optimizer node to remove first-sampling delay and improve CLIP text encoder warm-up

• v1.2.0 (2025-02-21)

• Added Model Optimized Loading node to optimize UNET model loading

• v1.1.0 (2025-02-20)

• Added Advanced Fusion node, supporting mixed processing for images and videos

• Added Simple Latent node

• v1.0.1 (2025-02-19)

• Added layer c