Supavector Agents Memory

Agent memory documentation for AI memory and vector retrieval

Learn how to create Memory, sync knowledge sources, call the /v1/memories APIs, and build AI memory workflows that work with LangChain or your own backend.

What this page covers

  • Covers Memory creation, search, ask, boolean_ask, chat, write, sync, and public chat routes.
  • Useful for developers building AI memory, vector retrieval, RAG, and grounded answer flows.
  • Explains how to move beyond a raw vector database by adding source sync, memory writes, public chat, and testing.

Common questions

Can Supavector work with LangChain?

Yes. Supavector can sit behind LangChain or your own runtime as the retrieval and durable memory layer while your application keeps its own orchestration and UX.

When should I use Supavector instead of Pinecone or Supabase pgvector alone?

Use Supavector when you need more than vector storage: source sync, grounded answers, memory-scoped APIs, public chat, and workflow-ready memory operations.