Graphiti

Graphiti

Build personalized AI agents that learn from dynamic data

416 followers

Graphiti is Knowledge Graph-based memory for AI agents. Automatically build rich graphs from changing business data & chat histories. Enable your Python agent with fast access to relevant, accurate data, even as it evolves over time. Visit our GitHub repo!
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Graphiti gallery image
Free
Launch Team / Built With
Migma
Migma
Lovable for Email
Promoted

What do you think? …

Danies Daniel
Very cool! What does zep do on top of graphiti?
Pavlo Paliychuk

@boatbuilder Thanks! Zep builds on Graphiti's temporal knowledge graph by providing a complete memory management layer. It introduces concepts such as user and group graphs and handles the infrastructural overhead of running Graphiti in your stack.

Vlad Arbatov

I'm genuinely impressed. @Graphiti might solve the long lived issue we've just discussed with the team in application to agentical memory being "not human enough".

Matt Wood

@vladzima Glad to hear it! We'd love to see how Graphiti could be of use to your dev team.

Marc Klingen

Congrats on the launch! We continue to see across @Langfuse users that memory is crucial for creating meaningful agents capable of engaging in long conversations and providing a personalized experience. Well done, Team Zep!

Matt Wood

@marc_klingenGreatly appreciated! A lot of work to be done to get agentic personalization right.

Danielle Blanchard

Ooh - this is wonderful!


I know how frustrating it is to get outdated info from platforms I'm engaging with.


Are you able to be quite specific about which data gets updated when using it on your app? If the AI agent is using multiple sources to generate information, can you decide which updated information is fed into Graphiti?