Fast, scalable building blocks for LLM Apps

Production-grade chat history memory, vector search, data enrichment, and more.

Quick Start Guide   |   Zep X LangChain 🦜⛓   |   Zep X LlamaIndex 🦙   |   Discord 💬
www.getzep.com

What is Zep?

Zep is an open source platform for productionizing LLM apps. Go from a prototype built in LangChain or LlamaIndex, or a custom app, to production in minutes without rewriting code.

Unlike vector databases, Users, Sessions, Memories, and Documents are all first-class citizens in Zep.

• Set up in 5 minutes

  ---

  Install via docker compose or Render and get up and running in minutes

  Quick Start

• Hybrid Vector Database

  ---

  Populate your prompts with relevant documents and chat history

  Vector Store

• Persist Chat History

  ---

  Store chat history, enrich it with metadata, and run semantic search over it

  Chat History

• Enrich it all

  ---

  Embeddings, summaries, named entities, token counts, and more

  Extractors

• Use the tools you love

  ---

  Python , TypeScript , LangChain 🦜⛓️, LlamaIndex 🦙, and more

  SDKs

• Open Source

  ---

  Zep is licensed under the Apache Software License and available on GitHub

  License

---

What you can build with Zep

1. Chatbot Applications: Zep's chat history storage enables context-aware chatbot applications.
2. Retrieval Augmented Generation (RAG) Applications: Zep's vector store aids in building applications that answer questions based on a document set.
3. Agent Applications: Zep's document vector store and enriched chat history store can be used as tools by agents, able to look up relevant information and past conversations.
4. Data Enrichment Applications: Zep's extractors can be used for applications that need to analyze and understand the text data from human/bot conversations.

---

Quick Start

Join the Zep Cloud Waitlist   - or -