n8n Voyage AI Embeddings Community Node

Developed and maintained by Jay Nguyen (Nguyễn Thiệu Toàn).

🛡️ Verified n8n Creator | 💼 CEO/Founder of GenStaff

Connect with me:

This is an n8n community node that integrates Voyage AI Embeddings into the n8n AI ecosystem (LangChain). It acts as a sub-node connected to Vector Store root nodes (such as Supabase, Pinecone, Qdrant, PGVector) to generate vector representations of text queries and documents.

Features

• Voyage 4 Models Support: Built-in options for voyage-4-large, voyage-4, voyage-4-lite, voyage-code-3, voyage-finance-2, and voyage-law-2.
• Zero Runtime Dependencies: Conforms to the strict n8n packaging guideline, ensuring no package collision or dependency bloat in your production deployment.
• Matryoshka Dimension Scaling: Custom dimension parameters (256, 512, 1024, 2048) to scale downstream database storage requirements.
• Output Quantization: Encodes outputs to float, int8, uint8, binary, or ubinary representation formats.
• Base64 Transfer Protocol: Decodes base64 payload response patterns for high-throughput network performance.
• Contextualized Truncation: Truncates inputs safely if they exceed context lengths.

Installation

For Self-Hosted n8n Instance

In your self-hosted n8n instance directory, run:

npm install n8n-nodes-embeddings-voyageai

Or use the Community Nodes menu in your n8n settings dashboard:

1. Go to Settings > Community Nodes.
2. Click Install a Node.
3. Type: n8n-nodes-embeddings-voyageai
4. Accept and install.

Local Development & Compilation

Requirements

• Node.js v18+ or v22+
• pnpm or npm

Steps

1. Clone this repository:

   git clone https://github.com/your-username/n8n-voyage-nodes-embedding.git
   cd n8n-voyage-nodes-embedding
   

2. Install dependencies:

   pnpm install
   

3. Build the node:

   pnpm run build
   

4. Link or load it in your local dev stack.

Local Verification using Docker Compose

A compose.yaml file is provided in the repository to mount your compiled output directory directly into a local n8n instance for testing:

1. Compile your TS files to JS:

   pnpm run build
   

2. Launch n8n using Docker Compose:

   docker compose up -d
   

3. Open http://localhost:5678 in your browser.
4. Navigate to the AI Canvas, add a Vector Store, and link the Voyage AI Embeddings node!

Author

• Jay Nguyen (Nguyễn Thiệu Toàn)
  • Verified n8n Creator: n8n Profile
  • CEO & Founder: GenStaff
  • Website: nguyenthieutoan.com
  • Email: me@nguyenthieutoan.com
  • LinkedIn: nguyenthieutoan
  • Facebook: nguyenthieutoan

License

MIT