Skip to content

voyage-3-large

Multilingual embeddings with long context and small dimensions

Visit official site

What voyage-3-large is

The model is a multilingual general-purpose embedding model from Voyage AI, described as ranking first across 100 datasets in eight domains such as law, finance and code.

What you can do with it

  • Retrieve legal cases and regulations in legal research
  • Analyze financial documents and reports to extract key information
  • Understand and generate code in software development

Getting started

  • Read the embeddings documentation on the Voyage AI site
  • Register and obtain an API key
  • Send text data and receive embedding vectors

Pros & cons

✓ What we like

  • Smaller dimensions and quantization lower vector database cost
  • The 32K token context exceeds OpenAI 8K and Cohere 512

! What to watch out for

  • An API key is required before calling the model
  • Lower precision quantization slightly changes retrieval quality

FAQ

What context length is supported?

The material states a 32K token context length.

How is the model called?

Register for an API key and send text data to the embeddings API.

Last reviewed: 2026-11-15

More AI coding tools tools

View all →

How we review