n8n-nodes-voyageai

This is an n8n community node package that provides VoyageAI embedding and reranking nodes for use in n8n workflows.

Installation

In your n8n instance, go to Settings > Community Nodes and install:

n8n-nodes-voyageai

Or install via npm:

npm install n8n-nodes-voyageai

Nodes

Embeddings VoyageAI

Generate text embeddings using VoyageAI's embedding models. Uses the official voyageai SDK directly.

Supported models:

| Model | Default Dimensions | Flexible Dimensions | Description | |-------|-------------------|---------------------|-------------| | voyage-4-large | 1024 | 256, 512, 1024, 2048 | Flagship model, best general-purpose and multilingual quality | | voyage-4 (default) | 1024 | 256, 512, 1024, 2048 | General-purpose model, strong quality | | voyage-4-lite | 1024 | 256, 512, 1024, 2048 | Optimized for latency and cost | | voyage-4-nano | 1024 | 256, 512, 1024, 2048 | Open-weight model, smallest and fastest | | voyage-code-3 | 1024 | 256, 512, 1024, 2048 | Optimized for code retrieval | | voyage-finance-2 | 1024 | - | Optimized for finance domain | | voyage-law-2 | 1024 | - | Optimized for legal domain | | voyage-multilingual-2 | 1024 | - | Optimized for multilingual content | | voyage-3.5 | 1024 | 256, 512, 1024, 2048 | Previous generation general-purpose | | voyage-3.5-lite | 1024 | 256, 512, 1024, 2048 | Previous generation lite | | voyage-3-large | 1024 | 256, 512, 1024, 2048 | Previous generation large |

Options: batch size, input type (query/document), output dimension (256/512/1024/2048), truncation, encoding format, output data type (float/int8/uint8/binary/ubinary).

Output dimension and output data type are shown conditionally for models that support flexible dimensions (voyage-4*, voyage-3.5*, voyage-3-large, voyage-code-3).

Embeddings VoyageAI Multimodal

Generate multimodal embeddings from text, images, or both using VoyageAI multimodal models.

Supported models:

| Model | Description | |-------|-------------| | voyage-multimodal-3.5 (default) | Latest multimodal model, supports text, images, and video screenshots | | voyage-multimodal-3 | Previous generation multimodal model |

Content types: Text only, Image URL only, Text + Image URL, Binary image only, Text + Binary image.

Embeddings VoyageAI Contextualized

Generate context-aware embeddings for document chunks using the voyage-context-3 model. Chunks with the same document ID are processed together to preserve inter-chunk relationships.

Required fields: Document ID field, Text field.

Reranker VoyageAI

Reorder documents by relevance to a query using VoyageAI's rerank models.

Known Limitation: NodeConnectionTypes.AiReranker does not currently work for n8n community nodes. This node is included for completeness but will not function until n8n lifts this restriction.

Supported models: rerank-2.5, rerank-2.5-lite, rerank-2, rerank-2-lite.

Credentials

This package requires a VoyageAI API credential with your API key.

1. Sign up at VoyageAI
2. Get your API key from the dashboard
3. In n8n, create a new VoyageAI API credential and paste your API key

Development

# Install dependencies
npm install

# Build
npm run build

# Run unit tests
npm run test:unit

# Run integration tests (requires VOYAGE_API_KEY)
VOYAGE_API_KEY=your-key npm run test:integration

# Run tests with coverage
npm run test:coverage

# Lint
npm run lint

License

MIT