@ragpipe/plugin-voyage

@ragpipe/plugin-voyage

Voyage AI embedding plugin for ragpipe.

This package provides the embedding axis only. Pair it with any ragpipe vector store and generation plugin.

Install

pnpm add ragpipe @ragpipe/plugin-voyage

Usage

import { defineConfig } from "ragpipe";
import { voyageEmbedding } from "@ragpipe/plugin-voyage";

export default defineConfig({
  embedding: voyageEmbedding({
    apiKey: process.env.VOYAGE_API_KEY!,
    model: "voyage-4-lite",
    dimensions: 1024,
    inputType: "document",
  }),
  // vectorStore, generation ...
});

API

voyageEmbedding(options)

Returns an EmbeddingPlugin that calls the Voyage embeddings API.

Option	Type	Default	Description
apiKey	string	—	Voyage API key (required)
model	string	—	Embedding model name (required)
dimensions	number	—	Output dimension (required)
baseUrl	string	"https://api.voyageai.com/v1"	API base URL
inputType	"query" | "document"	—	Retrieval-oriented input type hint

• Rate limit: 200ms between calls
• Batch support: embedMany() sends an array input in one API call
• Ordering: responses are sorted by index before returning

Supported Models

Common Voyage embedding model/dimension pairs:

Model	Dimensions
voyage-4-large	1024
voyage-4	1024
voyage-4-lite	1024
voyage-4-nano	1024
voyage-3-large	1024
voyage-3.5	1024
voyage-3.5-lite	1024
voyage-code-3	1024
voyage-code-2	1536
voyage-finance-2	1024
voyage-law-2	1024
voyage-large-2-instruct	1024
voyage-multilingual-2	1024

Notes

• Uses native fetch; no Voyage SDK dependency is required.
• Sends output_dimension on every request because dimensions is required.
• Sends input_type only when inputType is explicitly configured.
• Throws explicit errors for HTTP failures, malformed payloads, and empty embedding responses.

Get an API Key

• Sign in to Voyage AI
• Create an API key
• Set VOYAGE_API_KEY in your environment

License

MIT