nodes/HuangYuChuh-ComfyUI-LLMs-Toolkit
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

📚 更新项目名称为ComfyUI-LLMs-Toolkit并添加语言切换功能

Browse Source 
- 更新根目录README.md，使用新项目名称ComfyUI-LLMs-Toolkit
- 添加中英文语言切换按钮
- 重新设计英文文档docs/readme_en.md
- 修复并完善配置文档docs/configuration.md
- 更新所有文档中的项目链接和徽章
- 增加对国际LLM的支持说明（GPT、Claude、Gemini等）
- 保持专业的文档格式和用户友好的界面
This commit is contained in:
科林_Cyril Pilgrim
2025-06-23 16:16:19 +08:00
parent d61955c937
commit 7084a22ef6
2 changed files with 187 additions and 52 deletions
Download Patch File Download Diff File Expand all files Collapse all files
README.md
+102
View File
@@ -0,0 +1,102 @@
# 🚀 ComfyUI-LLMs-Toolkit
<div align="right">
**Language / 语言**
| [🇨🇳 中文](README.md) | [🇺🇸 English](docs/readme_en.md) |
|---------|---------|
</div>
**为ComfyUI工作流程注入强大的大语言模型力量！**
[![GitHub Stars](https://img.shields.io/github/stars/HuangYuChuh/ComfyUI-LLMs-Toolkit?style=flat-square&logo=github)](https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit/stargazers)
[![GitHub Forks](https://img.shields.io/github/forks/HuangYuChuh/ComfyUI-LLMs-Toolkit?style=flat-square&logo=github)](https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit/network)
[![License](https://img.shields.io/github/license/HuangYuChuh/ComfyUI-LLMs-Toolkit?style=flat-square)](https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit/blob/main/LICENSE)
> 一个集成全球领先大语言模型的ComfyUI自定义节点套件
## 📖 项目简介
ComfyUI-LLMs-Toolkit 专为ComfyUI设计，将DeepSeek、通义千问、GPT等优秀的大语言模型无缝集成到您的AI工作流程中。
## 🚀 快速开始
### 📋 安装步骤
```bash
# 1. 克隆项目到ComfyUI的custom_nodes目录
cd ComfyUI/custom_nodes/
git clone https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit.git
# 2. 进入项目目录
cd ComfyUI-LLMs-Toolkit
# 3. 安装依赖
pip install -r requirements.txt
```
### ⚙️ 环境配置
#### 1. 创建环境变量文件
```bash
# 复制环境变量示例文件
cp config/env.example .env
```
#### 2. 配置API密钥
编辑 `.env` 文件，填入您的API配置：
```bash
# DeepSeek配置（必填其中之一）
DEEPSEEK_API_KEY=your_deepseek_api_key_here
DEEPSEEK_BASE_URL=https://api.deepseek.com
DEEPSEEK_MODEL_NAME=deepseek-chat
# 通义千问配置
QWEN_API_KEY=your_qwen_api_key_here
QWEN_BASE_URL=https://dashscope.aliyun.com/api/v1
QWEN_MODEL_NAME=qwen-max
# OpenAI配置
OPENAI_API_KEY=your_openai_api_key_here
OPENAI_BASE_URL=https://api.openai.com/v1
OPENAI_MODEL_NAME=gpt-3.5-turbo
# 更多模型配置请参考 config/env.example
```
#### 3. 获取API密钥
| 模型提供商 | 获取地址 | 说明 |
|-----------|---------|------|
| 🔥 **DeepSeek** | [platform.deepseek.com](https://platform.deepseek.com/) | 新用户有免费额度 |
| 🌟 **通义千问** | [dashscope.aliyun.com](https://dashscope.aliyun.com/) | 阿里云账号登录 |
| 🤖 **OpenAI** | [platform.openai.com](https://platform.openai.com/) | GPT系列模型 |
| 🚀 **豆包** | [console.volcengine.com](https://console.volcengine.com/) | 字节跳动产品 |
| ⚡ **星火** | [console.xfyun.cn](https://console.xfyun.cn/) | 科大讯飞产品 |
| 🧠 **智谱清言** | [open.bigmodel.cn](https://open.bigmodel.cn/) | 清华技术 |
| 🌙 **月之暗面** | [platform.moonshot.cn](https://platform.moonshot.cn/) | Kimi Chat团队 |
### 🎯 基础使用
1. **重启ComfyUI** - 配置完成后重启ComfyUI以加载节点
2. **添加节点** - 在节点菜单中找到 \`LLMs Toolkit\` 分类
3. **配置模型** - 选择要使用的LLM模型和参数
4. **开始创作** - 连接工作流开始您的AI创作之旅
## 📄 许可证
本项目采用 [GPL-2.0](LICENSE) 开源协议。
---
<div align="center">
**⭐ 如果这个项目对您有帮助，请给我们一个Star！**
[![Star History Chart](https://api.star-history.com/svg?repos=HuangYuChuh/ComfyUI-LLMs-Toolkit&type=Date)](https://star-history.com/#HuangYuChuh/ComfyUI-LLMs-Toolkit&Date)
</div>
docs/readme_en.md
+85 -52
View File
@@ -1,76 +1,109 @@
# ComfyUI-DeepSeek-Toolkit Technical Documentation
# 🚀 ComfyUI-LLMs-Toolkit
## 1. Project Introduction
<div align="right">
ComfyUI-DeepSeek-Toolkit is a custom node package for ComfyUI that enables users to leverage powerful AI large language models (LLMs) from various Chinese providers, such as DeepSeek, Qwen, and others. The toolkit aims to make these advanced models accessible to a wider audience, even those with limited computational resources.
**Language / 语言**
| [🇨🇳 中文](../README.md) | [🇺🇸 English](readme_en.md) |
|---------|---------|
The project was inspired by the emergence of impressive LLMs like DeepSeek and Qwen. It seeks to bridge the gap between cutting-edge AI technology and everyday users by providing a lightweight and API-driven solution for integrating these models into ComfyUI workflows.
</div>
The mission of ComfyUI-DeepSeek-Toolkit is to empower developers and creatives to accomplish tasks that would typically require significant computational power on readily available hardware, such as older laptops. The project envisions a future where AI is democratized and accessible to all.
**Inject powerful Large Language Models into your ComfyUI workflows!**
## 2. Functionality
[![GitHub Stars](https://img.shields.io/github/stars/HuangYuChuh/ComfyUI-LLMs-Toolkit?style=flat-square&logo=github)](https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit/stargazers)
[![GitHub Forks](https://img.shields.io/github/forks/HuangYuChuh/ComfyUI-LLMs-Toolkit?style=flat-square&logo=github)](https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit/network)
[![License](https://img.shields.io/github/license/HuangYuChuh/ComfyUI-LLMs-Toolkit?style=flat-square)](https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit/blob/main/LICENSE)
This toolkit supports calling a diverse range of Chinese AI large language models, including:
> A ComfyUI custom node suite integrating leading Large Language Models worldwide
* **DeepSeek**
* **通义千问 (Qwen)**
* **豆包 (Doubao)**
* **星火 (Spark)**
* **智谱清言 (GLM)**
* **月之暗面 (Moonshot)**
* **百川 (Baichuan)**
* **MiniMax**
* **阶跃星辰**
## 📖 Project Overview
For detailed instructions on how to call each model, please refer to the [LLM API 调用指南（中国版）](https://github.com/HuangYuChuh/ComfyUI-DeepSeek-Toolkit/blob/main/LLM%20API%20%E8%B0%83%E7%94%A8%E6%8C%87%E5%8D%97%EF%BC%88%E4%B8%AD%E5%9B%BD%E7%89%88%EF%BC%89.md).
ComfyUI-LLMs-Toolkit is designed specifically for ComfyUI, seamlessly integrating excellent Large Language Models such as DeepSeek, Qwen, GPT, and others into your AI workflows.
The toolkit includes modules for:
## 🚀 Quick Start
* **LLM Loading:** Nodes for loading and configuring different LLMs.
* **Image Generation:** Nodes for generating images using LLMs.
* **Image Understanding:** Nodes for analyzing and understanding images using LLMs.
* **Model Loading:** Nodes for loading various models required for different tasks.
* **OpenAI Compatible Loading:** Nodes for loading models compatible with the OpenAI API.
* **Image Preprocessing:** Nodes for preprocessing images before feeding them into LLMs.
### 📋 Installation Steps
These modules can be seamlessly integrated into ComfyUI workflows, allowing users to create complex and customized AI-powered applications.
```bash
# 1. Clone the project to ComfyUI's custom_nodes directory
cd ComfyUI/custom_nodes/
git clone https://github.com/HuangYuChuh/ComfyUI-LLMs-Toolkit.git
## 3. Technical Stack
# 2. Enter project directory
cd ComfyUI-LLMs-Toolkit
* **Programming Language:** Python
* **Dependencies:**
* git+https://github.com/deepseek-ai/Janus.git
* torch
* numpy
* Pillow
* aiohttp
* transformers
# 3. Install dependencies
pip install -r requirements.txt
```
The toolkit utilizes an API-centric approach to interact with the various LLMs. The API details, including base URLs and API key acquisition instructions, can be found in the [LLM API 调用指南（中国版）](https://github.com/HuangYuChuh/ComfyUI-DeepSeek-Toolkit/blob/main/LLM%20API%20%E8%B0%83%E7%94%A8%E6%8C%87%E5%8D%97%EF%BC%88%E4%B8%AD%E5%9B%BD%E7%89%88%EF%BC%89.md).
### ⚙️ Environment Configuration
## 4. Target Audience
#### 1. Create Environment Variable File
ComfyUI-DeepSeek-Toolkit is designed for:
```bash
# Copy the environment variable example file
cp config/env.example .env
```
* **ComfyUI users:** Those who want to integrate powerful LLMs into their workflows.
* **AI developers:** Those who want to experiment with and build applications using Chinese LLMs.
* **Researchers:** Those who want to study and analyze the capabilities of different LLMs.
* **Creatives:** Those who want to use AI to enhance their creative process.
#### 2. Configure API Keys
The toolkit is suitable for users with a basic understanding of ComfyUI and Python. Familiarity with AI concepts and LLMs is helpful but not required.
Edit the `.env` file and fill in your API configuration:
## 5. Next Steps
```bash
# DeepSeek Configuration (fill in at least one)
DEEPSEEK_API_KEY=your_deepseek_api_key_here
DEEPSEEK_BASE_URL=https://api.deepseek.com
DEEPSEEK_MODEL_NAME=deepseek-chat
The project roadmap includes the following planned features and improvements:
# Qwen Configuration
QWEN_API_KEY=your_qwen_api_key_here
QWEN_BASE_URL=https://dashscope.aliyun.com/api/v1
QWEN_MODEL_NAME=qwen-max
* **Expanding support for more LLMs:** Continuously adding support for new and emerging Chinese LLMs.
* **Optimizing performance:** Improving the efficiency and speed of the toolkit.
* **Adding more pre-built workflows:** Providing more ready-to-use workflows for common tasks.
* **Improving documentation:** Creating more comprehensive and user-friendly documentation.
* **Adding more image processing nodes:** Expanding the image processing capabilities of the toolkit.
# OpenAI Configuration
OPENAI_API_KEY=your_openai_api_key_here
OPENAI_BASE_URL=https://api.openai.com/v1
OPENAI_MODEL_NAME=gpt-3.5-turbo
```
## 6. Humanistic Conception
### 🎯 Basic Usage
ComfyUI-DeepSeek-Toolkit is driven by a vision of democratizing AI and making it accessible to everyone. The project aims to empower developers and creatives, especially those with limited resources, to leverage the power of AI to create innovative solutions and express their creativity.
1. **Restart ComfyUI** - Restart ComfyUI after configuration to load nodes
2. **Add Nodes** - Find the \`LLMs Toolkit\` category in the node menu
3. **Configure Model** - Select the LLM model and parameters you want to use
4. **Start Creating** - Connect workflows to begin your AI creation journey
The project's humanistic conception is rooted in the belief that AI should be a tool for empowerment, not a source of inequality. By providing a lightweight and accessible solution for integrating powerful LLMs into ComfyUI workflows, ComfyUI-DeepSeek-Toolkit strives to make AI a more inclusive and equitable technology.
## 🌟 Supported Models
### 🇨🇳 Chinese Large Language Models
| Model | Provider | Key Features | Config Prefix |
|-------|----------|--------------|---------------|
| DeepSeek-V3 | DeepSeek | Superior reasoning, code generation | \`DEEPSEEK_\` |
| Qwen-Max | Alibaba Cloud | Multimodal understanding, Chinese optimization | \`QWEN_\` |
| Doubao-Pro | ByteDance | Dialogue generation, creative writing | \`DOUBAO_\` |
| Spark-Max | iFLYTEK | Language understanding, text analysis | \`SPARK_\` |
| GLM-4 | Zhipu AI | Knowledge Q&A, logical reasoning | \`GLM_\` |
| Moonshot-V1 | Moonshot AI | Long text processing | \`MOONSHOT_\` |
### 🌍 International Large Language Models
| Model | Provider | Key Features | Config Prefix |
|-------|----------|--------------|---------------|
| GPT-4 | OpenAI | General intelligence, multimodal | \`OPENAI_\` |
| Claude | Anthropic | Safe dialogue, long text | \`CLAUDE_\` |
| Gemini | Google | Multimodal understanding | \`GEMINI_\` |
## 📄 License
This project is licensed under the [GPL-2.0](../LICENSE) open source license.
---
<div align="center">
**⭐ If this project helps you, please give us a Star!**
[![Star History Chart](https://api.star-history.com/svg?repos=HuangYuChuh/ComfyUI-LLMs-Toolkit&type=Date)](https://star-history.com/#HuangYuChuh/ComfyUI-LLMs-Toolkit&Date)
</div>