
Graphiti
β Text source-checked Nov 17, 2026Open-source framework for building temporal knowledge graphs that power Zep.
Source-checked against the vendor's site Β· rebuilt from the vendor's own pagesProfile updated Link checked How we rate
In short
Graphiti is an open-source framework for building temporal knowledge graphs that power Zep. It ingests data as episodes to create dynamic, time-aware graphs for LLM applications.
Graphiti is an open-source framework for building temporal knowledge graphs, known as Context Graphs, that power Zep. Graphiti incrementally updates graphs as new episodes arrive, allowing relationships and facts to evolve with the source data. Graphiti builds dynamic, temporally-aware knowledge graphs that represent complex, evolving relationships between entities over time.
Graphiti ingests both unstructured and structured data, and the resulting graph may be queried using a fusion of time, full-text, semantic, and graph algorithm approaches. Graphiti's hybrid search combines vector similarity, BM25 full-text, and graph traversal into a single ranked answer without LLM-in-the-loop reranking. Graphiti processes data as discrete episodes, maintaining data provenance and allowing incremental entity and relationship extraction. Graphiti supports defining domain-specific entity types, enabling more precise knowledge representation for specialized applications.
Graphiti enables the construction of LLM applications such as assistants that learn from user interactions and fuse personal knowledge with dynamic data from business systems. Graphiti supports agents that autonomously execute complex tasks by reasoning with state changes from multiple dynamic sources. Graphiti supports applications in sales, customer service, health, and finance, enabling long-term recall and state-based reasoning for assistants and agents.
Graphiti runs on Neo4j, FalkorDB, or Amazon Neptune, with LLM and embedding providers including OpenAI, Azure OpenAI, Gemini, and Anthropic. The Graphiti MCP Server is an experimental implementation that exposes Graphiti operations through the Model Context Protocol. MCP clients can use the server to add episodes, search the graph, manage graph data, and inspect the server status.
According to Graphiti: Graphiti is an open-source framework for building temporal knowledge graphs, Known as Context Graphs, that power Zep. Its pages list this strength: Incremental graph updates as new episodes arrive.
β The AI Tool Bible editorial team
Pros
- β Incremental graph updates as new episodes arrive.
- β Temporal awareness tracks changes in facts and relationships over time.
- β Hybrid search combines vector similarity, BM25 full-text, and graph traversal.
- β Open-source framework.
- β MCP server allows clients to add episodes, search the graph, and manage data.
Facts
- Platforms
- api (not source-checked)
- API available
- Yes (not source-checked)
- Free trial
- Yes (not source-checked)
- Open source
- Yes
- Model used
- Multi-model (not source-checked)
Where this tool sits
Use cases
Frequently asked
- Is Graphiti free to use?
- Yes, Graphiti is an open-source framework. The pricing is listed as free, though no specific details are provided regarding potential costs for hosting or associated LLM providers.
- Which database backends does Graphiti support?
- Graphiti runs on Neo4j, FalkorDB, or Amazon Neptune. It also supports various LLM and embedding providers, including OpenAI, Azure OpenAI, Gemini, and Anthropic, allowing for flexible infrastructure choices.
- How does Graphiti handle data updates?
- Graphiti processes data as discrete episodes, allowing it to incrementally update graphs as new information arrives. This approach maintains data provenance and enables relationships and facts to evolve dynamically with the source data.
- What search capabilities does Graphiti offer?
- Graphiti uses a hybrid search approach that combines vector similarity, BM25 full-text, and graph traversal. This fusion of time, full-text, semantic, and graph algorithms produces a single ranked answer without requiring LLM-in-the-loop reranking.
- Can Graphiti be used with MCP clients?
- Yes, the Graphiti MCP Server is an experimental implementation that exposes Graphiti operations through the Model Context Protocol. MCP clients can use it to add episodes, search the graph, manage data, and inspect server status.