Loading tool details…
Temporal knowledge graph memory service that gives AI agents accurate, time-aware recall of users, business data, and past conversations.
Temporal knowledge graph memory service that gives AI agents accurate, time-aware recall of users, business data, and past conversations.
Zep is a memory platform for LLM agents built around Graphiti, its open-source temporal knowledge graph engine. Instead of appending raw chat history to prompts, Zep continuously extracts entities, relationships, and events from conversations and business data, then stores them in a bi-temporal graph that knows both when a fact was true and when the system learned it. That lets agents answer questions like "what plan is this customer on right now?" or "what did they promise us last quarter?" without confusing outdated snapshots with current state. Zep integrates with LangChain, LlamaIndex, LangGraph, CrewAI, and the OpenAI Agents SDK; it also supports ingestion of business data (CRM records, support tickets, documents) so agent memory unifies conversational and system context. Facts include provenance so agents can cite sources. Zep provides per-user summarization, semantic search, entity linking, and rate-limited retrieval to keep contexts small. Pricing (from getzep.com/pricing, July 2026): Free developer tier, Pro tier starting at $25/month with $75 tier options, and Enterprise with on-prem, HIPAA, and SOC 2. Graphiti itself is Apache 2.0 and self-hostable.
Zep delivers sophisticated context engineering capabilities that go far beyond simple conversation memory. Users praise the temporal knowledge graph approach for capturing entity relationships and fact evolution over time. The <200ms retrieval latency and framework-agnostic integration make it suitable for real-time applications. Enterprise features including SOC2 and HIPAA compliance address security requirements. Some users note the credit-based pricing can become expensive at scale, and the graph-based architecture requires more setup than simple memory stores.
Facts carry validity windows. When new information contradicts old, Zep invalidates the old fact rather than deleting it — the agent reasons on the current truth but you can audit history.
Benchmarked at 148ms for 1K graphs and 168ms for 100M graphs — retrieval stays flat as your Context Lake grows.
Access control, retention, legal hold, and audit live in the underlying platform, not glued onto each agent — critical for compliance at enterprise scale.
Expose a Zep memory space as an MCP server; any MCP-compatible client can call it as a first-class memory tool.
$0
$0
$25/month
$75/month
Contact sales
Ready to get started with Zep?
View Pricing Options →Zep works with these platforms and services:
We believe in transparent reviews. Here's what Zep doesn't handle well:
Pricing and features change often. If anything on this Zep page is wrong or missing, tell us and include a link to the source. We review every correction.