Memory & RAG

Give AI memory across conversations, isolate it per user, and ground answers in your own documents with embeddings, vector stores, and RAG.

💭 Memory

Memory lets an AI keep context across turns. BoxLang AI ships more than 20 memory types, and every one supports multi-tenant isolation with userId and conversationId.

Kind	Types	Use it for
Conversation	Windowed, Summary, Session, File, Cache, JDBC	Chat history and context
Vector	BoxVector, Chroma, Postgres, MySQL, OpenSearch, TypeSense, Pinecone, Qdrant, Weaviate, Milvus	Semantic search and RAG
Hybrid	Recent plus semantic	Best of both

Isolate memory per user and per conversation:

chat = aiMemory(
    memory: "window",
    key: createUUID(),
    userId: "user123",
    conversationId: "support-ticket-456",
    config: { maxMessages: 10 }
)

Multi-tenant isolation is built in, so one customer's conversation never leaks into another's.

📚 RAG

Retrieval-Augmented Generation grounds answers in your own documents. It reduces hallucinations, keeps knowledge current without retraining, and tracks which sources informed an answer.

A complete RAG system in a few lines:

// 1. Create vector memory
vectorMemory = aiMemory( memory: "chroma", config: {
    collection: "knowledge_base",
    embeddingProvider: "openai",
    embeddingModel: "text-embedding-3-small"
} )

// 2. Ingest documents
aiDocuments( "/docs", {
    type: "directory",
    recursive: true,
    extensions: [ "md", "txt", "pdf" ]
} ).toMemory(
    memory = vectorMemory,
    options = { chunkSize: 1000, overlap: 200 }
)

// 3. Create an agent that retrieves automatically
agent = aiAgent(
    name: "Knowledge Assistant",
    instructions: "Answer using the provided documentation. If unsure, say so.",
    memory: vectorMemory
)

println( agent.run( "How do I create a custom loader?" ) )

📥 Document Loaders

Load content from text, Markdown, HTML, CSV, JSON, XML, PDF, logs, HTTP, feeds, SQL queries, whole directories, and web crawlers. Loaders chunk documents and send them straight to memory.

🔍 Go Deeper

• Memory Systems
• Vector Memory
• Multi-Tenant Memory
• Embeddings
• Document Loaders
• RAG

Use Couchbase as a vector memory backend with the Couchbase + AI Memory module. Every BoxLang+ module is free to try for 60 days.