Zep

What is Zep?

Zep is a managed context layer that gives AI agents memory of their users and the data around them. Developers send messages and other data to Zep Cloud, which builds temporal context graphs with the open-source Graphiti framework, then returns a Context Block of facts, entities, summaries and observations when the agent needs it. User graphs and standalone graphs are kept fully isolated from each other, and a standalone graph can be used to share context across users. Zep Cloud is accessed through Python, TypeScript and Go SDKs, a dashboard with graph visualization and logs, and webhooks on some plans. Governance features include audit logs, API logs, policy-based access and customer-key encryption, and the company lists SOC 2 Type II and HIPAA compliance. Deployment options are Cloud, bring your own keys, or bring your own cloud. Zep has a free tier with a monthly episode-credit limit; the vendor does not publish paid prices, so check its pricing page.

Use cases

• Give a chatbot long-term memory of each user
• Build a context graph from chat history and business data
• Retrieve a Context Block for each agent prompt
• Share context across users with a standalone graph
• Track customer and account context for support agents
• Handle group-chat memory in one graph
• Get notified when ingestion finishes through webhooks

Who it's for

• AI engineers
• Agent developers
• Product teams
• Enterprise AI teams

Why use Zep

• Agents get memory that updates as users and data change, rather than a fixed document index.
• Isolated per-user graphs keep one customer's information away from another's.
• A Context Block assembled for you reduces the retrieval code you maintain.
• A free tier lets you try the managed service before paying.
• Enterprise teams can choose a managed cloud or run Zep in their own cloud with BYOC.

Zep features

• Temporal knowledge graph memory powered by Graphiti
• Context Block with facts, entities, episodes, observations and summaries
• User graphs and standalone graphs, isolated from one another
• Python, TypeScript and Go SDKs
• Dashboard with graph visualization, debug logs and API logs
• Webhooks for episode processed and batch completed events
• Bring your own LLM credentials with rate-limit alerts
• Audit logs, API logs, policy-based access and customer-key encryption
• SOC 2 Type II and HIPAA compliance
• Cloud, BYOK and BYOC deployment

Zep integrations

• Graphiti
• Claude Code
• Codex
• Cursor

How to get started with Zep

1. Sign up for Zep Cloud and install the Python, TypeScript or Go SDK.
2. Create users and add messages or business data; episodes are processed asynchronously into a graph.
3. At prompt time, retrieve a Context Block for the user's thread.
4. Pass the Context Block to your model along with the new message.
5. Use the dashboard, logs and webhooks to watch ingestion and debug retrieval.

Zep limitations

• Zep Community Edition is deprecated and no longer supported
• The free tier has a limited number of episode credits per month
• Webhooks are available only on certain plans
• Official multilingual support is still on the roadmap
• Designed for changing data, not as a replacement for RAG over static documents

Zep pricing

Yes, Zep has a free plan.

Prices can change and may vary by country or exclude tax. Check the official pricing page.

Zep alternatives

The main Zep alternatives listed on Debuly are Mem0, Letta, Cognee, LangMem, Supermemory and Hindsight. The table compares pricing and platforms; order is A–Z.

1. 

   Cognee

   Open-source memory engine for AI agents that combines vector search, knowledge graphs and relational storage, with a hosted cloud.

   Developer platformsAutomation and agentsResearch and search
   FreemiumVisit Cognee (official site)
2. 

   Hindsight

   MIT-licensed memory engine from Vectorize that lets AI agents retain, recall and learn from experience, self-hosted or as a managed cloud.

   Developer platformsAutomation and agents
   Open sourceVisit Hindsight (official site)
3. 

   LangMem

   Python library from LangChain that gives AI agents long-term memory and lets them refine behavior from past conversations.

   Developer platformsAutomation and agents
   Open sourceVisit LangMem (official site)
4. 

   Letta

   Platform, formerly MemGPT, for building stateful AI agents with persistent memory, from a terminal app, SDK or API.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Letta (official site)
5. 

   Mem0

   Memory layer for AI agents and apps, available as an open-source package or a managed platform that keeps user context across sessions.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Mem0 (official site)
6. 

   Supermemory

   Memory and context API for AI agents with user profiles, a memory graph, retrieval and connectors, billed by usage.

   Developer platformsAutomation and agentsCoding
   FreemiumVisit Supermemory (official site)

Zep FAQ

Sources & verification

Facts on this page come from the official sources below, each with the date we read it. Written by Debuly; the vendor did not review this page.

Image credits

• Logo: from help.getzep.com · fetched 24 Oct 2026 · Vendor trademark, used only to identify the product; removed on request
• Cover: none collected yet, so a generated tile is shown.

Spot an error? Report an issue with this listing. We also remove a logo or image on the owner's request.