n8n-nodes-volcengine-ai-enhanced

This is an enhanced n8n community node for VolcEngine Ark (火山方舟) AI services, forked and improved from crazyyanchao/n8n-nodes-volcengine-ai.

🙏 Acknowledgments

This project is based on the excellent work by @crazyyanchao. We extend our sincere thanks for the original implementation.

Original Repository: crazyyanchao/n8n-nodes-volcengine-ai

✨ Key Enhancements

This enhanced version introduces the following major improvements:

🚀 Dynamic Endpoint Loading

• Automatic Endpoint Discovery: Dynamically loads available endpoints from VolcEngine Ark API instead of hardcoded model lists
• Real-time Updates: Always shows the latest available endpoints with "Running" status
• Smart Filtering: Automatically filters and displays only active endpoints

🧠 Thinking Mode Support

• Deep Thinking Mode: Full support for DeepSeek R1, DeepSeek V3.2, and other models with thinking capabilities
• Three Modes Available:
  • disabled: Force disable deep thinking
  • enabled: Force enable deep thinking with reasoning chain output
  • auto: Model automatically decides when to use deep thinking
• Official API Compliance: Implements according to VolcEngine Ark documentation

🔐 Enhanced Authentication

• Dual Authentication Support:
  • API Token for model API calls (Bearer or X-Api-Access-Key)
  • AK/SK (Access Key/Secret Key) for ListEndpoints API with HMAC-SHA256 signature
• VolcEngine V4 Signature: Proper implementation of VolcEngine API signature algorithm

🎯 Focused on Chat Model

• Chat Model Node Only: This project focuses exclusively on the chat model node for n8n's AI workflows
• n8n Integration: Designed specifically for use with:
  • AI Chain: Connect to AI Chain nodes to build complex reasoning workflows
  • AI Agent: Connect to AI Agent nodes for autonomous task execution with reasoning capabilities
  • LangChain Compatibility: Fully compatible with n8n's LangChain integration

📋 Features

This node provides:

• Dynamic Endpoint Selection: Choose from live endpoints loaded from VolcEngine Ark
• Chat Completion: Intelligent conversation capabilities with multi-turn support
• Thinking Mode: Enable deep reasoning for complex tasks
• Rich Parameter Configuration: Temperature, max tokens, penalties, response format, etc.
• LangChain Integration: Compatible with n8n's AI chain and agent workflows

🚀 Installation

Follow the installation guide in the n8n community nodes documentation.

Using npm

npm install n8n-nodes-volcengine-ai-enhanced

Manual Installation

1. Clone this repository
2. Run npm install
3. Run npm run build
4. Copy the dist folder to your n8n custom nodes directory

🔑 Credentials Setup

To use this node, you need to configure the VolcEngine AI Enhanced API credentials:

Required Fields

1. Access Token: Your VolcEngine AI API token (for model API calls)

   • Used for chat completion requests
   • Supports both Bearer and X-Api-Access-Key authentication

2. Access Key ID (AK): Your VolcEngine Access Key ID

   • Required for ListEndpoints API calls
   • Used for HMAC-SHA256 signature authentication

3. Secret Access Key (SK): Your VolcEngine Secret Access Key

   • Required for ListEndpoints API calls
   • Used for HMAC-SHA256 signature authentication

4. Region: VolcEngine region (default: cn-beijing)

   • Used for API endpoint selection

How to Get Credentials

1. Register VolcEngine Account: Visit VolcEngine Official Website to register
2. Enable Ark Service: Enable VolcEngine Ark service in the console
3. Create API Token: Create an API token in the VolcEngine console for model API calls
4. Create Access Key: Create an Access Key (AK/SK) in the VolcEngine console for ListEndpoints API

📖 Usage Instructions

Basic Workflow

1. Install the Node: Install this community node in your n8n instance
2. Configure Credentials:
   • Add "VolcEngine AI Enhanced API" credentials
   • Fill in Access Token, Access Key ID, Secret Access Key, and Region
3. Add Node to Workflow:
   • Add "VolcengineAi Chat Model (Enhanced)" node to your workflow
   • Connect it to AI Chain or AI Agent nodes
4. Select Endpoint:
   • Choose an endpoint from the dropdown (dynamically loaded)
   • Or manually enter an endpoint ID
5. Configure Options (optional):
   • Enable Thinking Mode if your model supports it
   • Adjust temperature, max tokens, etc.
6. Execute Workflow: Run the workflow to get AI responses

Thinking Mode Configuration

For models that support deep thinking (e.g., DeepSeek R1, DeepSeek V3.2):

1. Select your endpoint
2. In Options, set Thinking Mode to:
   • disabled: No thinking chain output
   • enabled: Always output thinking chain
   • auto: Let the model decide

Example: Using with AI Chain

[AI Chain] → [VolcengineAi Chat Model (Enhanced)] → [Output]

The enhanced node provides the language model to the AI Chain, which can then process complex workflows with reasoning capabilities.

⚙️ Available Options

• Base URL: Override default API endpoint
• Frequency Penalty: Reduce repetitive content (-2 to 2)
• Max Retries: Maximum retry attempts
• Maximum Number of Tokens: Max output length (up to 32768)
• Presence Penalty: Encourage new topics (-2 to 2)
• Reasoning Effort: Control reasoning token usage (low/medium/high)
• Response Format: Text or JSON output
• Sampling Temperature: Control randomness (0-2)
• Thinking Mode: Enable/disable/auto deep thinking
• Timeout: Request timeout in milliseconds
• Top P: Nucleus sampling diversity (0-1)

🔧 Compatibility

• Minimum n8n Version: 1.0.0
• Node.js Version: >=22.16
• Tested Version: n8n 1.0.0+

📚 Resources

• n8n Community Nodes Documentation
• VolcEngine Ark Documentation
• VolcEngine Ark API Reference
• VolcEngine Ark Deep Thinking Guide
• VolcEngine Official Website

📝 License

MIT License - Same as the original project

🤝 Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

⚠️ Notes

• This node requires a valid VolcEngine account with Ark service enabled
• Please ensure compliance with VolcEngine's terms of use and API call limits
• The ListEndpoints API requires AK/SK authentication
• Thinking mode is only available for models that support it (e.g., DeepSeek R1, DeepSeek V3.2)

Forked from: crazyyanchao/n8n-nodes-volcengine-ai
Maintained by: @foxtwobao