← Back to Directory

Voyage AI

LLM Orchestrators

Overview

High-performance embedding models specifically designed for RAG and information retrieval. Voyage AI's models consistently top benchmarks for retrieval accuracy and domain-specific knowledge handling.

Voyage AI (now part of MongoDB) builds high-performance embedding and reranking models tuned for retrieval accuracy, including domain-specific variants for code, finance, and law. Upgrading embeddings is often the fastest way to improve a RAG system, and Voyage's models top many benchmarks. It targets teams pushing retrieval quality.

Key Features

  • State-of-the-art embedding models
  • Reranking models for retrieval
  • Domain-specific variants
  • Simple API
  • Strong benchmark performance

Best For

Teams that want to maximize retrieval accuracy in RAG systems.

Pros & Cons

Pros
  • Top-tier retrieval accuracy
  • Domain-specialized models
  • Easy to adopt
Cons
  • Embeddings only, not full stack
  • Usage-based costs
Advertisement

Pulse Verdict

The retrieval precision engine. If your RAG system's accuracy is hitting a wall, Voyage AI's specialized embeddings are the most efficient way to break through.

Pricing

Usage-based API pricing with a free allowance.

Pricing changes often — confirm current plans on the official site.

Visit Official Website →

Related Tools

Jina AI Reader

A powerful API that converts any URL into high-quality, LLM-friendly markdown. It is designed to provide clean, structured text for RAG pipelines and autonomous agents.

LlamaIndex

A data framework for LLM applications that provides powerful tools for ingesting, structuring, and accessing private or domain-specific data. It is the leading library for building complex RAG systems.