Welcome to vai docs
The vai documentation site is live. This site provides comprehensive documentation for voyageai-cli, the CLI tool for Voyage AI embeddings and MongoDB Atlas Vector Search.
What you will find hereโ
Getting Started covers installation, quickstart, configuration, and shell completions.
Core Concepts explains the building blocks: embeddings, vector search, RAG, reranking, cosine similarity, and more. These are the same explanations available via vai explain <topic> in your terminal.
Command Reference documents all 33 vai commands with synopsis, options, and examples.
Guides walk through real workflows:
- Building a RAG pipeline in 5 minutes
- Chatting with your documents using LLMs
- Composing multi-step workflows as JSON
- Setting up the MCP server in Claude Desktop or Cursor
- Benchmarking and evaluating retrieval quality
API Reference covers environment variables, MCP tool parameters, and the workflow JSON schema.
Getting startedโ
npm install -g voyageai-cli
vai quickstart
Or dive into the installation guide.
Contributingโ
This site is open source at github.com/mrlynn/vai-docs. Found an error? Every page has an "Edit this page" link at the bottom.
