Voyage AI Provider for Vercel AI SDK

Introduction

The voyage-ai-provider is a community provider for the Vercel AI SDK that enables seamless integration with Voyage AI's powerful embedding models. It simplifies the process of generating text embeddings for various applications, including semantic search, recommendation systems, and clustering.

What are Embeddings?

Embeddings are numerical representations of text that capture semantic meaning. They are essential for tasks that require understanding the relationships between words and phrases. Voyage AI provides state-of-the-art embedding models that can generate high-quality embeddings for various use cases.

Why Use Voyage AI with Vercel AI SDK?

The Vercel AI SDK is a popular sdk for building AI-powered applications. By integrating Voyage AI's embedding models, you can enhance your applications with advanced natural language understanding capabilities. The voyage-ai-provider makes this integration straightforward and efficient.

Installation

Install the voyage-ai-provider using your preferred package manager:

npm install voyage-ai-provider
# or
yarn add voyage-ai-provider
# or
pnpm add voyage-ai-provider
# or
bun add voyage-ai-provider

Configuration

To use the Voyage AI Provider, you need an API key from Voyage AI. Sign up at Voyage to obtain your key.

Add your API key to your .env file:

.env

VOYAGE_API_KEY=your-api-key

Usage

Basic Usage

Import the default provider instance voyage and use the .textEmbeddingModel() method to create an embedding model:

ai.ts

import { voyage } from 'voyage-ai-provider';
import { embedMany } from 'ai';

const embeddingModel = voyage.textEmbeddingModel('voyage-3-lite');

export const generateEmbeddings = async (
  value: string,
): Promise<Array<{ embedding: number[]; content: string }>> => {
  // Generate chunks from the input value
  const chunks = value.split('\n');

  // Optional: You can also split the input value by comma
  // const chunks = value.split('.');

  // Or you can use LLM to generate chunks(summarize) from the input value

  const { embeddings } = await embedMany({
    model: embeddingModel,
    values: chunks,
  });
  return embeddings.map((e, i) => ({ content: chunks[i], embedding: e }));
};

Advanced Usage: Customizing the Provider Instance

If you need a customized setup, import createVoyage and create a provider instance with your settings:

ai.ts

import { createVoyage } from 'voyage-ai-provider';

const voyage = createVoyage({
  baseURL: 'https://api.voyageai.com/v1', // Optional: Customize the base URL
  apiKey: process.env.VOYAGE_API_KEY,
  // Optional: Add custom headers
});

Adding Settings to the Model

You can pass additional settings to the embedding model during initialization. Refer to the Voyage API documentation for available settings.

ai.ts

const embeddingModel = voyage.textEmbeddingModel('voyage-3-lite', {
  inputType: 'document',
  outputDimension: '1024', // For voyage-code-3, you can choose from 256, 512, 1024 (default), 2048
  outputDtype: 'float',
});

Voyage Embedding Models

Voyage AI offers a variety of embedding models, each optimized for different use cases:

Model	Context Length (tokens)	Embedding Dimension
voyage-3	32,000	1024
voyage-3-lite	32,000	512
voyage-code-3	32,000	1024 (default), 256, 512, 2048
voyage-finance-2	32,000	1024
voyage-multilingual-2	32,000	1024
voyage-law-2	16,000	1024
voyage-code-2	16,000	1536

Community Providers

The voyage-ai-provider is a community-maintained provider for the Vercel AI SDK. It is developed and maintained by patelvivekdev .

Author

patelvivekdev

Voyage AI Provider for Vercel AI SDK