Find tools › Open-source projects › Chat with your files

graphiti

Open-source project · by getzep

A Python toolkit that turns conversations and data into a map of facts that remembers when each was true.

For developersPopular888,911 installs a month

What it costs

Free and open source; you may pay for the AI model it uses.

Use it to

  • Track how facts about customers change over time
  • Give an AI agent memory of people, things and events
  • Search connected facts instead of plain text

How to start

  1. Set up a graph database such as Neo4j and get an LLM API key.

  2. In a terminal, run pip install graphiti-core.

  3. Follow the README quickstart to add conversations and search the resulting graph.

Good to know. You must run a graph database yourself (Neo4j, FalkorDB or Amazon Neptune) and supply an LLM API key. The README notes that production features like dashboards, user management and tuned performance are in the paid Zep service. The speed comparisons with GraphRAG are the project's own.

Before you install

  • Checked by a personWe read what it does before listing it.
  • License: Apache-2.0You are allowed to use it under these terms.
  • Last updated 2026-11-17

More detail

A Python framework that builds a knowledge graph from conversations and data, recording facts as connections between people, things and ideas, along with when each fact was true.

Use it when an agent needs memory that handles changing facts (for example, 'lived in Paris until 2025') and can trace each fact back to its source, rather than just searching text chunks. It is the open-source core of Zep's commercial product.

Similar tools

Descriptions are written in plain English from each project's own documentation and refreshed by our agent; this one was last checked 2026-09-29. Spotted a mistake? · Open in Find tools · © 2026 Which AI Map