n8n-nodes-volcengine-ark

This is an n8n community node that provides two ways to use VolcEngine ARK (Doubao) in your n8n workflows:

1. VolcEngine ARK - Regular node for direct API calls
2. VolcEngine ARK Chat Model - LangChain-compatible chat model for AI Agents

VolcEngine ARK provides powerful large language models (LLM) through the ARK platform.

n8n is a fair-code licensed workflow automation platform.

Installation

Follow the installation guide in the n8n community nodes documentation.

Community Nodes (Recommended)

1. Go to Settings > Community Nodes
2. Select Install
3. Enter n8n-nodes-volcengine-ark in Enter npm package name
4. Agree to the risks and select Install

Manual Installation

To get started install the package in your n8n root directory:

npm install n8n-nodes-volcengine-ark

For Docker-based deployments add the following line before the font installation command in your n8n Dockerfile:

RUN cd /usr/local/lib/node_modules/n8n && npm install n8n-nodes-volcengine-ark

Credentials

To use this node, you need to create credentials in n8n:

1. API Key: Get your API key from the VolcEngine ARK console
2. Base URL: Default is https://ark.cn-beijing.volces.com/api/v3 (usually no need to change)

Nodes

VolcEngine ARK (Regular Node)

Direct API integration for chat completions and image generation.

Operations

Chat

Generate a chat completion with Doubao models.

Parameters:

• Resource: Chat
• Operation: Complete
• Model: Select from predefined models or enter custom model ID
• Messages: Array of messages with role (system/user/assistant) and content
• Options (optional):
  • Temperature: Controls randomness (0-2)
  • Top P: Nucleus sampling parameter (0-1)
  • Max Tokens: Maximum tokens to generate
  • Stream: Enable streaming responses
  • Frequency Penalty: Reduce repetition (-2 to 2)
  • Presence Penalty: Encourage new topics (-2 to 2)
  • Stop Sequences: Stop generation at specific sequences

Embedding (Text & Multimodal Vectorization)

Generate vector embeddings for text and images for semantic search, similarity matching, and RAG applications.

Parameters:

• Resource: Embedding
• Operation:
  • Create Text Embedding: Convert text to vectors
  • Create Multimodal Embedding: Convert text and/or images to vectors
• Model: Select embedding model (e.g., doubao-embedding-large-text-250515)
• Input Type (Text Embedding):
  • Single Text: Embed one text
  • Multiple Texts (Batch): Embed multiple texts efficiently
• Content Type (Multimodal Embedding):
  • Text Only: Embed text content
  • Image Only: Embed image from URL
  • Text + Image: Cross-modal embedding
• Options (optional):
  • Encoding Format: Float (default) or Base64
  • Dimensions: Custom vector dimensions (1-2048, default: 2048)
  • User ID: Optional user identifier for tracking

Use Cases:

• 🔍 Semantic search and information retrieval
• 📊 Text classification and clustering
• 🤝 Content similarity matching
• 🖼️ Image-to-text and text-to-image search
• 💬 RAG (Retrieval-Augmented Generation) systems

See Embedding Usage Guide for detailed examples and best practices.

Image (Image Generation)

Generate images from text prompts or reference images using VolcEngine's Seedream models.

Parameters:

• Resource: Vision
• Operation: Generate Image
• Model Source: Choose from predefined models or enter custom model ID
• Model: Select image generation model (e.g., Seedream 4.0, Seedream 3.0 T2I)
• Prompt: Text description of the image you want to generate
• Reference Images (optional): URLs of reference images (one per line) for image-to-image or multi-image generation
• Additional Parameters (JSON): Advanced options such as:
  • size: Image dimensions (e.g., "1024x1024")
  • watermark: Whether to add watermark (true/false)
  • response_format: Response format ("url" or "b64_json")
  • seed: Random seed for reproducibility
  • guidance_scale: How closely to follow the prompt (1-20)
  • sequential_image_generation: Enable sequential generation
  • stream: Enable streaming responses

Example Additional Parameters:

{
  "size": "1024x1024",
  "watermark": true,
  "response_format": "url",
  "seed": 42,
  "guidance_scale": 7.5
}

Supported Models:

• Seedream 4.0 (250828): Latest generation model supporting text-to-image, image-to-image, and multi-image generation
• Seedream 3.0 T2I (250415): Text-to-image model with various size options

VolcEngine ARK Chat Model (AI Agent Compatible)

LangChain-compatible chat model node designed for use with n8n AI Agents, AI Tools, and AI Memory nodes.

Features

• Seamless integration with n8n AI Agents
• Support for multi-turn conversations
• Full LangChain compatibility
• Reasoning model support (DeepSeek R1, Doubao Thinking Pro)

Parameters

• Model: Choose from all available VolcEngine ARK models
• Options:
  • Temperature: Controls randomness (0-2, default: 0.7)
  • Maximum Tokens: Max tokens to generate (default: 4096)
  • Top P: Nucleus sampling (0-1)
  • Frequency Penalty: Reduce repetition (-2 to 2)
  • Presence Penalty: Encourage new topics (-2 to 2)
  • Stop Sequences: Up to multiple stop sequences
  • Timeout: Request timeout in milliseconds (default: 60000)
  • Max Retries: Number of retry attempts (default: 2)

Usage Examples

Using the Regular Node

1. Add the VolcEngine ARK node to your workflow
2. Connect it to your trigger
3. Configure credentials
4. Select Chat resource and Complete operation
5. Choose your model from the list or enter a custom model ID
6. Add messages (e.g., system prompt and user question)
7. Execute the workflow

Using the Chat Model with AI Agent

1. Add an AI Agent node to your workflow
2. In the AI Agent settings, select VolcEngine ARK Chat Model as the language model
3. Configure the model selection and parameters
4. Connect AI Tools and AI Memory nodes as needed
5. The Chat Model will automatically handle conversation context and tool calling

Supported Models

• Doubao Seed 1.6 Series: Flagship models with ultimate speed
• Doubao 1.5 Series: Pro and Lite versions with 32k/256k context
• Doubao Thinking Pro: Deep reasoning model
• DeepSeek Series: V3 and R1 models with distilled variants

Compatibility

• Tested with n8n version 1.0.0+
• Compatible with VolcEngine ARK API v3
• LangChain integration via @langchain/openai
• Works with n8n AI Agent, AI Tool, and AI Memory nodes

Resources

• n8n community nodes documentation
• VolcEngine ARK documentation

License

MIT