ComfyUI——BOZO Plugin Collection

|

Introduction

This plugin collection includes multiple practical nodes covering JSON processing, image generation, translation, LLM dialogue, image segmentation, OSS storage, HTML/Markdown processing, and audio synthesis functions. Suitable for ComfyUI workflow processes, it enhances creative efficiency.

Environment Configuration Detection Method

Install Python libraries from requirements.txt using pip. You can first execute python check_requirements.py to check if dependencies are installed completely.

---

Installation Steps

1. Clone the project and place this plugin folder in ComfyUI's custom_nodes directory.

    git clone https://github.com/bozoyan/comfyui_bozo.git

2. Install dependencies:

    pip install -r requirements.txt

3. Restart the ComfyUI service. The new nodes will be visible in ComfyUI's node panel.

Node Categories and Functions

1. Example Nodes (Learning Purposes)

• a1基础格式
  a1
  Example: Demonstrates basic node format structure.
• a2基础数据类型
  a2
  Example: Demonstrates different data type inputs and outputs.
• a3基础调用流程
  a3
  Example: Shows basic calling logic between nodes.
• a4一个可以运行的节点
  a4
  Complete example: A runnable node for testing basic functions.
• a5最简格式
  a5
  Minimalist node implementation template.

---

2. Computational Nodes

Basic Operations

• 比较数值~ 🎯BOZO
  CompareInt
  Compares the size relationship between two integers.
• 规范数值~ 🎯BOZO
  FloatToInteger
  Converts floating point numbers to integers.
• 生成范围数组~ 🎯BOZO
  GenerateNumbers
  Generates a sequence of numbers within a specified range.
• 范围内随机数~ 🎯BOZO
  GetRandomIntegerInRange
  Generates a random integer within a specified range.

---

3. String Processing Nodes

Basic Text Processing

• 文本输入~ 🎯BOZO
  SingleTextInput
  Single-line text input node.
• 文本到列表~ 🎯BOZO
  TextToList
  Splits text into a list by lines.
• 文本拼接~ 🎯BOZO
  TextConcatenator
  Concatenates multiple text strings.
• 多参数输入~ 🎯BOZO
  MultiParamInputNode
  Input node supporting multiple parameter types.
• 整数参数~ 🎯BOZO
  NumberExtractor
  Extracts integer parameters from text.
• 添加前后缀~ 🎯BOZO
  AddPrefixSuffix
  Adds prefix and suffix to text.
• 提取标签之间~ 🎯BOZO
  ExtractSubstring
  Extracts substring between two tags.
• 按数字范围提取~ 🎯BOZO
  ExtractSubstringByIndices
  Extracts substring based on start and end indices.
• 分隔符拆分两边~ 🎯BOZO
  SplitStringByDelimiter
  Splits string into two parts using a delimiter.
• 常规处理字符~ 🎯BOZO
  ProcessString
  Performs regular processing operations on strings.
• 提取前后字符~ 🎯BOZO
  ExtractBeforeAfter
  Extracts content before or after a specified string.
• 简易文本替换~ 🎯BOZO
  SimpleTextReplacer
  Simple text replacement function.
• 替换第n次出现~ 🎯BOZO
  ReplaceNthOccurrence
  Replaces the nth occurrence of specified content in a string.
• 多次出现依次替换~ 🎯BOZO
  ReplaceMultiple
  Replaces multiple occurrences of strings sequentially.
• 批量替换字符~ 🎯BOZO
  BatchReplaceStrings
  Batch replaces multiple strings.
• 随机行内容~ 🎯BOZO
  RandomLineFromText
  Randomly selects a line from text.
• 判断是否包含字符~ 🎯BOZO
  CheckSubstringPresence
  Checks if text contains a specified substring.
• 段落每行添加前后缀~ 🎯BOZO
  AddPrefixSuffixToLines
  Adds prefix and suffix to each line in a paragraph.
• 段落提取指定索引行~ 🎯BOZO
  ExtractAndCombineLines
  Extracts lines at specified indices from a paragraph and combines them.
• 段落提取或移除字符行~ 🎯BOZO
  FilterLinesBySubstrings
  Filters lines based on whether they contain substrings.
• 段落字数条件过滤行~ 🎯BOZO
  FilterLinesByWordCount
  Filters text lines based on word count conditions.
• 按序号提取分割文本~ 🎯BOZO
  SplitAndExtractText
  Extracts split text by specified sequence number.
• 文本出现次数~ 🎯BOZO
  CountOccurrences
  Counts occurrences of a substring in text.
• 文本拆分~ 🎯BOZO
  ExtractLinesByIndex
  Splits text by index.
• 提取特定行~ 🎯BOZO
  ExtractSpecificLines
  Extracts specific lines from text.
• 删除标签内的内容~ 🎯BOZO
  RemoveContentBetweenChars
  Deletes content between specified tags.
• 随机打乱~ 🎯BOZO
  ShuffleTextLines
  Randomly shuffles the order of text lines.
• 判断返回内容~ 🎯BOZO
  ConditionalTextOutput
  Returns different text content based on conditions.
• 文本按条件判断~ 🎯BOZO
  TextConditionCheck
  Checks if text meets specified conditions.
• 文本组合~ 🎯BOZO
  TextConcatenation
  Combines multiple text strings.
• 提取多层指定数据~ 🎯BOZO
  ExtractSpecificData
  Extracts specified data from multi-layer structures.
• 指定字符行参数~ 🎯BOZO
  FindFirstLineContent
  Finds the first line containing specified characters and extracts parameters.
• 获取整数~ 🎯BOZO
  GetIntParam
  Gets integer parameters.
• 获取浮点数~ 🎯BOZO
  GetFloatParam
  Gets floating point parameters.
• 视频指令词模板~ 🎯BOZO
  GenerateVideoPrompt
  Generates video prompt templates.

---

4. File Processing Nodes

Image Processing

• 加载图像（按大小）~ 🎯BOZO
  LoadAndAdjustImage
  Loads and adjusts images according to specified size.
• 加载图像（按路径）~ 🎯BOZO
  GenericImageLoader
  Loads images via file path.
• 调整图像（比例）~ 🎯BOZO
  ImageAdjuster
  Adjusts image proportions and dimensions.
• 裁剪图像（宽高）~ 🎯BOZO
  CustomCrop
  Crops images by specified width and height.
• 保存图像（本地）~ 🎯BOZO
  SaveImagEX
  Saves images to local storage.
• 文件操作~ 🎯BOZO
  FileCopyCutNode
  File copy or move operations.
• 替换文件名~ 🎯BOZO
  FileNameReplacer
  Replaces specified content in filenames.
• 文本写入TXT~ 🎯BOZO
  WriteToTxtFile
  Writes text to TXT files.
• 清理文件~ 🎯BOZO
  FileDeleteNode
  Deletes specified files.
• 加载文件（路径列表）~ 🎯BOZO
  FileListAndSuffix
  Gets file path list.
• 图像层叠加~ 🎯BOZO
  ImageOverlayAlignment
  Overlays one image onto another.
• 文字图像~ 🎯BOZO
  TextToImage
  Converts text to images.

Spreadsheet Processing

• 读取表格数据~ 🎯BOZO
  ReadExcelData
  Reads data from Excel files.
• 写入表格数据~ 🎯BOZO
  WriteExcelData
  Writes data to Excel files.
• 图片插入表格~ 🎯BOZO
  WriteExcelImage
  Inserts images into Excel files.
• 查找表格数据~ 🎯BOZO
  FindExcelData
  Finds data in Excel files.
• 读取表格数量差~ 🎯BOZO
  ReadExcelRowOrColumnDiff
  Reads Excel row/column quantity differences.

---

5. Functional Nodes

Basic Functions

• 当前时间(戳)~ 🎯BOZO
  GetCurrentTime
  Gets current timestamp.
• 随机整数~ 🎯BOZO
  SimpleRandomSeed
  Generates random integer seed.
• 选择参数~ 🎯BOZO
  SelectionParameter
  Parameter selector.
• 读取页面~ 🎯BOZO
  ReadWebNode
  Reads web page content.
• 解码预览~ 🎯BOZO
  DecodePreview
  Decodes and previews images.

JSON Processing Nodes

Basic Operations

• JSON解析器~ 🎯BOZO
  B_JSONParserNode
  Parses JSON strings into Python dictionaries.

• Input:

  • json_string: JSON string
  • path: Data path (optional)

• Output:

  • parsed_data: Parsed JSON or specific value
  • array_size: Length when result is an array

• JSON 随机值~ 🎯BOZO
  B_RandomJSONValueNode
  Randomly selects a value from a JSON object.

• Input:

  • json_input: JSON string
  • max_depth: Maximum depth for random selection

• Output:

  • random_value: Randomly selected value

• JSON 合并~ 🎯BOZO
  B_JSONMergeNode
  Merges two or more JSON objects.

• Input:

  • json_input_1: First JSON
  • json_input_2: Second JSON
  • merge_strategy: Merge strategy (overwrite/preserve/concatenate)

• Output:

  • merged_json: Merged JSON

• JSON 修改~ 🎯BOZO
  B_JSONModifierNode
  Modifies values in JSON objects based on paths.

• Input:

  • json_input: JSON to be modified
  • path: Modification path
  • new_value: New value

• Output:

  • modified_json: Updated JSON

• JSON 字典~ 🎯BOZO
  B_JSONGeneratorNode
  Generates standard JSON dictionaries from input.

• Input:

  • key_value_pairs: Key-value pairs
  • is_array: Whether to generate an array instead of an object

• Output:

  • generated_json: New JSON structure

Advanced Operations

• JSON 长度~ 🎯BOZO
  B_JSONLengthNode
  Gets the length of JSON arrays or objects.

• Input:

  • json_input: JSON input

• Output:

  • length: Number of items

• JSON Key值~ 🎯BOZO
  B_JSONKeyCheckerNode
  Checks if a specified key exists.

• Input:

  • json_input: JSON object
  • key: Key to check

• Output:

  • exists: Boolean result
  • value: Value when it exists

• JSON 字符串生成~ 🎯BOZO
  B_JSONStringifierNode
  Converts JSON objects to formatted strings.

• Input:

  • json_input: JSON input
  • indent: Number of indentation spaces
  • sort_keys: Whether to sort keys alphabetically

• Output:

  • json_string: Formatted JSON string

Iteration and Traversal

• JSON Object对象~ 🎯BOZO
  B_JSONObjectIteratorNode
  Traverses key-value pairs of JSON objects.

• Input:

  • json_input: JSON object
  • index: Current index
  • mode: Iteration mode (fixed/increment/decrement)

• Output:

  • key: Current key
  • value: Current value
  • current_index: Current position
  • total_items: Total items

• JSON Array数组~ 🎯BOZO
  B_JSONArrayIteratorNode
  Traverses elements of JSON arrays.

• Input:

  • json_input: JSON array
  • index: Current index
  • mode: Iteration mode (fixed/increment/decrement)

• Output:

  • item: Current item
  • current_index: Current position
  • total_items: Total items

Path Syntax

• Nested objects: object.nestedObject.property
• Array elements: array[0] or array.0
• Complex structures: object.array[2].property

1. JSON Traversal

# Input JSON
{
    "users": [
        {"name": "Alice", "age": 30},
        {"name": "Bob", "age": 25}
    ]
}

# Using array iterator, mode="incr"
# Will output each user object sequentially

2. Merging JSON Objects

# First input
{"name": "John", "age": 30}

# Second input
{"age": 31, "city": "New York"}

# Result using overwrite strategy
{"name": "John", "age": 31, "city": "New York"}

3. Random Value Selection

# Input JSON
{
    "colors": ["red", "blue", "green"],
    "sizes": {"S": 10, "M": 20, "L": 30}
}

# May return any color or size value

Image Processing Nodes

• 新建噪点图片~ 🎯BOZO
  Bozo_Pic
  Generates noise images of specified dimensions.

• 自定义Latent尺寸~ 🎯BOZO
  BOZO_Custom_Image
  Adjusts the dimensions of latent space images.

• 图片增强GPEN~ 🎯BOZO
  BOZO_GpenImage
  Enhances image quality using ModelScope's GPEN model.

• 图片放大GPEN~ 🎯BOZO
  B_GpenImage
  Enlarges images using the GPEN model.

• 图片运算~ 🎯BOZO
  BImageYunSuan
  Performs basic mathematical operations on images.

• 图片区域对比~ 🎯BOZO
  B_quyu
  Compares image regions.

• 图片羽化~ 🎯BOZO
  B_yuhua
  Feathering processing on images.

• 图片渐变透明化~ 🎯BOZO
  B_touming
  Adds gradient transparency effects to images.

• 图片合成合并~ 🎯BOZO
  B_hebin
  Synthesizes and merges multiple images.

• 图片URL反推（ModelScope）~ 🎯BOZO
  ImageJiexi
  Parses image information from URLs.

• 文生图（ModelScope）~ 🎯BOZO
  PicRun
  Text-to-image generation.

• 保存图像（URL）~ 🎯BOZO
  BImageSave
  Saves images to specified URLs.

• 保存图片元数据~ 🎯BOZO
  BImageSaveWithExtraMetadata
  Saves images with additional metadata.

• 图片预览~ 🎯BOZO
  PreviewPic
  Previews generated images.

• 单图片浏览器~ 🎯BOZO
  PicChrome
  Views single images in Chrome browser.

• 多图片浏览器~ 🎯BOZO
  PicSChrome
  Views multiple images in Chrome browser.

• 加载图像（多图）~ 🎯BOZO
  B_ImagesInput
  Supports input of multiple images.

• 加载图像（批量）~ 🎯BOZO
  Bozo_ImagesInput
  Supports batch input of images.

• 加载图像（网络）~ 🎯BOZO
  ImagePathLoader
  Loads images from network paths.

• 图片读取器~ 🎯BOZO
  ImageLoader
  Reads image files.

• 元数据读取器~ 🎯BOZO
  PNGInfoReader
  Reads metadata of PNG images.

• 元数据提取器~ 🎯BOZO
  PNGInfoExtractor
  Extracts specific metadata from PNG images.

Text and Language Processing Nodes

• 百度翻译~ 🎯BOZO
  ComfyUI_FanYi
  Performs text translation using Baidu Translation API (requires API key configuration).

• API接口~ 🎯BOZO
  BOZO_LLMAPI_NODE
  Universal LLM API calling interface.

• LLM文本提示词~ 🎯BOZO
  BOZO_LLM_Node
  Engages in text dialogue with large language models.

• 思考型LLM~ 🎯BOZO
  BOZO_Node
  Engages in thinking-style dialogue using ModelScope's Qwen model, outputting thought processes and final answers.

• 字符串合并拼接~ 🎯BOZO
  Bhebin
  Merges multiple strings.

• 文本预览~ 🎯BOZO
  Bozo_preview_text
  Previews text content.

• 文本MD读取~ 🎯BOZO
  BOZO_TXT_MD
  Reads Markdown text.

• 调用文本行数据~ 🎯BOZO
  Bozo_SplitNode
  Extracts specific lines from multi-line text.

• 输出OS系统变量~ 🎯BOZO
  BozoPrintOS
  Prints operating system environment variables.

• GLM文本提示词~ 🎯BOZO
  GLM_Text_Chat
  Engages in text dialogue using GLM model.

• GLM识图生成提示词~ 🎯BOZO
  GLM_Vision_ImageToPrompt
  Generates prompts through images using GLM model.

• GLM文本翻译~ 🎯BOZO
  GLM_Translation_Text
  Performs text translation using GLM model.

HTML and Markdown Processing Nodes

• Markdown转HTML~ 🎯BOZO
  MarkmapNode
  Converts Markdown to HTML format.

• 读取Markdown~ 🎯BOZO
  ReadHtmlNode
  Reads Markdown files.

• HTML转图片~ 🎯BOZO
  HtmlToImageNode
  Converts HTML content to images.

• 保存Markdown~ 🎯BOZO
  BozoSaveMd
  Saves Markdown content to files.

• 保存HTML~ 🎯BOZO
  BozoSaveHTML
  Saves HTML content to files and optionally opens in browser.

Alibaba Cloud Image Segmentation Nodes

Notes

• First configure Alibaba Cloud Image Segmentation Service, obtain AccessKey ID and AccessKey Secret.
• Save the keys to the plugin directory's AssetKey.json file (refer to example for format).

Node List

• 衣服Seg分割-Cloth~ 🎯BOZO
  ALY_Seg_Cloth
  Performs image segmentation on clothing.

• 物体Seg分割-Obj~ 🎯BOZO
  ALY_Seg_Obj
  Performs image segmentation on general objects.

• 头部Seg分割-head~ 🎯BOZO
  ALY_Seg_head
  Performs image segmentation on head regions.

• 皮肤Seg分割-Skin~ 🎯BOZO
  ALY_Seg_Skin
  Performs image segmentation on skin regions.

Cloud Storage and API Nodes

• OSS上传器~ 🎯BOZO
  OSSUploader
  Uploads files to Alibaba Cloud OSS storage.

• OSS链接输出~ 🎯BOZO
  OSSUrlOutput
  Generates and outputs access links for OSS files.

• 文字推理~ 🎯BOZO-X
  X_API_Node
  Performs text reasoning using X platform API.

• 图片推理~ 🎯BOZO-X
  X_API_Image
  Performs image reasoning using X platform API.

• 图片生成~ 🎯BOZO-X
  X_API_Images
  Generates images using X platform API.

• Gemini 单图生成~ 🎯BOZO
  Google-Gemini
  Generates images using Google Gemini model.

SiliconFlow Nodes

• 文生图~ 🎯BOZO SiliconFlow
  BOZO_SiliconFlow_Txt2Img
  Converts text to images using SiliconFlow service.

• LLM对话~ 🎯BOZO SiliconFlow
  BOZO_SiliconFlow_LLM
  Engages in dialogue using SiliconFlow's LLM service.

• 图像分析~ 🎯BOZO SiliconFlow
  BOZO_SiliconFlow_ImageAnalysis
  Analyzes image content using SiliconFlow service.

• JSON代码生成~ 🎯BOZO SiliconFlow
  BOZO_SiliconFlow_JSONGenerator
  Generates JSON data using SiliconFlow service.

Audio Processing Nodes

• 音频-上传Base64~ 🎯BOZO
  BOZO_SiliconFlow_Audio_UploadBase64
  Uploads Base64 encoded audio data.

• 音频-复刻声音~ 🎯BOZO
  BOZO_SiliconFlow_Audio_UploadFile
  Uploads audio files for voice cloning.

• 音频-音色列表~ 🎯BOZO
  BOZO_SiliconFlow_Audio_ListVoices
  Gets list of available voices.

• 音频-删除音色~ 🎯BOZO
  BOZO_SiliconFlow_Audio_DeleteVoice
  Deletes custom voices.

• 音频-自定义音色~ 🎯BOZO
  BOZO_SiliconFlow_Audio_CustomVoice
  Synthesizes speech using custom voices.

• 音频-系统音色~ 🎯BOZO
  BOZO_SiliconFlow_Audio_SystemVoice
  Synthesizes speech using system preset voices.

• 音频-文件筛选~ 🎯BOZO
  BOZO_SiliconFlow_Audio_FileSelector
  Filters audio files.

• 音频-文字转录~ 🎯BOZO
  BOZO_SiliconFlow_Audio_FilePicker
  Selects audio files and performs text transcription.

Other Utility Nodes

• 加载CSV~ 🎯BOZO
  StylesCSVLoader
  Loads and parses style data from CSV files.

• Qwen-Image LoRA转换器~ 🎯BOZO
  B_QwenLoraConverterNode
  Converts ModelScope-trained Qwen-Image LoRA files to ComfyUI recognizable format.

• Nunchaku LoRA转换器~ 🎯BOZO
  B_NunchakuLoraConverterNode
  Nunchaku LoRA converter.

• 多图分析~ 🎯BOZO
  B_KontextDuoImageAnalyzer
  Multi-image analysis.

• 豆包生图~ 🎯BOZO
  DoubaoImageGenerator
  Doubao image generation.

• 豆包改图~ 🎯BOZO
  B_DoubaoImageEdit
  Doubao image editing.

• 豆包文生图SeeDream~ 🎯BOZO
  volcengine-seedream-v3
  Doubao text-to-image SeeDream.

• 豆包文生视频T2V~ 🎯BOZO
  volcengine-t2v
  Doubao text-to-video T2V.

• 豆包图片编辑ImgEditV3~ 🎯BOZO
  volcengine-img-edit-v3
  Doubao image editing ImgEditV3.

• 豆包首尾帧视频Seedance~ 🎯BOZO
  volcengine-doubao-seedance
  Doubao first-and-last-frame video Seedance.

---

KEY Configuration Instructions

Alibaba Cloud Image Plugin Key Configuration

Login to Alibaba Cloud Node: Currently supports clothing + objects

https://vision.aliyun.com/imageseg?spm=api-workbench.api_explorer.0.0.604ce85cenrYuD

Obtain key and fill in key/AssetKey.json

1. Create Alibaba Cloud AssetKey.json file in the plugin key directory.
2. Example content:

   {
     "access_key_id": "your_access_key_id Access Key ID for identifying user identity.",
     "access_key_secret": "your_access_key_secret Access Key Secret for verifying user identity."
   }

LLM API Key Configuration

1. Create corresponding API key files in the plugin key directory, just put the key text directly:
   • ModelScope API: modelscope_api_key.txt
   • Siliconflow API: siliconflow_API_key.txt
   • Other API key files

Baidu Translation API Configuration

1. Create AssetKey_Baidu.json file in the plugin key directory.
2. Example content:

   {
     "appid": "your_appid Baidu Translation API appid",
     "appkey": "your_appkey Baidu Translation API appkey"
   }

OSS Storage Configuration

1. Create AssetKey_OSS.json file in the plugin key directory.
2. Example content:

    {
      "access_key_id": "your_access_key_id Access Key ID for identifying user identity.", 
      "access_key_secret": "your_access_key_secret Access Key Secret for verifying user identity.",
      "bucket_name": "your_bucket_name Your selected Bucket name",
      "oss_domain": "your_oss_domain OSS access address, recommended to configure as https://",
      "directory": "your_directory/ Relative path to save OSS files in Bucket, needs trailing slash",
      "name_prefix": "your_prefix File name prefix"
    }

olcengine (Volcengine) API Configuration

1. Create Volcengine.json file in the plugin key directory.
2. Example content:

    {
      "api_key": "your_api_key Volcengine API Key",
      "model": "your_model_name Text-to-image model name",
      "model_edit": "your_edit_model_name Image editing model name",
      "model_id": "your_model_id Image understanding model ID",
      "base_url": "your_base_url Volcengine base URL"
    }

GLM Configuration

1. Create glm.json file in the plugin key directory (if not already created).
2. Example content:

    {
      "ZHIPUAI_API_KEY": "your_zhipu_api_key Zhipu AI API key",
      "from_translate": "zh Source language, default Chinese",
      "to_translate": "en Target language, default English",
      "model": "your_model_name Model name"
    }

---

Simple Usage Examples

1. Text-to-Image Generation Workflow

Text Input -> BOZO LLM Text Dialogue -> Text-to-Image -> Image Enhancement GPEN -> Image Save

2. Image Segmentation and Processing Workflow

Image Loader -> Object Seg Segmentation -> Image Operations -> Save Image with Metadata

3. Multimodal Content Generation and OSS Upload Workflow

Text Input -> SiliconFlow LLM Dialogue -> Markdown to HTML -> HTML to Image -> OSS Uploader -> OSS Link Output

4. Audio Synthesis Workflow

Text Input -> Audio-System Voice -> Audio File Selection -> Audio-Voice Cloning -> Audio-Custom Voice

---

Update Log

• 2024.12: Initial version released, including basic JSON processing and image processing nodes
• 2025.01: Added Alibaba Cloud image segmentation and OSS storage functions
• 2025.02: Added LLM dialogue and HTML/Markdown processing functions
• 2025.03: Added SiliconFlow nodes and audio processing functions
• 2025.04: Optimized documentation and user interface, added more examples
• 2025.05: Added Baidu Translation API function and other LLM interface support
• 2025.06: Fixed audio file upload issues, added ModelScope text-to-image node functions, added other utility node functions
• 2025.07: Added cloud model API support for GLM, Grok, Qwen, etc., fixed empty line and line break issues in dialogue, added streaming reply function for dialogue
• 2025.08: BizyAIR cloud API nodes development
• 2025.09: Short Video API - Cloud Platform

---

📬 Contact and Support

• Issues: Submit Issues

Thank you for using comfyui_bozo! 🎉

The related calls and GUI page functions of the project are still being continuously improved...

Hope this tool brings convenience to you. 😊

---

Star History

---

License

MIT License