Zep / Graphiti logo

Zep / Graphiti

Free

Build Temporal Context Graphs for AI Agents

FreeFree tier
Inputs: text
Type
Open Source
Company
Zep

About Zep / Graphiti

Graphiti is an open-source framework for building and querying temporal context graphs for AI agents. Unlike static knowledge graphs, Graphiti tracks how facts change over time, maintains provenance to source data, and supports both prescribed and learned ontology. It continuously integrates user interactions, structured and unstructured enterprise data, and external information into a coherent, queryable graph. The framework supports incremental data updates, efficient retrieval, and precise historical queries without requiring complete graph recomputation, making it suitable for developing interactive, context-aware AI applications. Graphiti's context graphs contain entities (nodes) with evolving summaries, temporal facts/relationships (edges) with validity windows, and episodes (provenance) as raw ingested data.

Key Features

Temporal knowledge graphs that track how facts change over time
Provenance to source data via episodes
Supports both prescribed and learned ontology
Incremental data updates without full graph recomputation
Hybrid retrieval: semantic + keyword + graph traversal
Autonomous graph building from unstructured and structured data
MCP server support for Claude, Cursor, and other MCP clients

Pros & Cons

Pros
  • Temporal awareness: tracks validity windows for facts, enabling historical queries
  • Provenance tracking: each fact traces back to source data (episodes)
  • Supports both prescribed and learned ontology for flexibility
  • Hybrid retrieval combining semantic, keyword, and graph traversal for rich context
  • Open-source and free to use
  • Incremental updates avoid costly full recomputation
Cons
  • Requires self-hosting and infrastructure setup for production use
  • May have a learning curve for teams unfamiliar with knowledge graphs or temporal data models

Best For

Building context graphs that evolve with every interactionProviding AI agents with rich, structured context instead of flat document chunks or raw chat historyQuerying across time, meaning, and relationships with hybrid retrievalTracking what is true now and what was true before for evolving data