---
title: "past.dev vs Zep"
description: "Compare Zep and past.dev by temporal model, database, ingestion format, deployment, and recall output."
canonical: https://past.dev/vs/zep
last-updated: 2026-10-10
---
# past.dev vs Zep

Source: https://past.dev/vs/zep

Zep uses Graphiti, an open-source temporal graph engine that runs on a dedicated graph database. past.dev provides a JSON API for timestamped source ingestion and grouped, ranked recall with dated source excerpts. It runs as a managed service or, on an Enterprise agreement, in your own cloud. Choose between an engine you operate and an API contract your application calls.

## What Zep is

Zep is a managed memory service for agents built on [Graphiti](https://github.com/getzep/graphiti), an Apache 2.0 framework maintained by the same team. Graphiti ingests episodes and derives entity nodes and relationship edges. Each fact has a validity window and a link to its source episode. New information can invalidate an earlier fact while retaining it for history. Graphiti runs on Neo4j, FalkorDB, and Amazon Neptune. The team published the architecture as a paper on temporal knowledge graphs for agent memory.

## What past.dev is

past.dev accepts raw text and its original timestamp without a required schema. The same public routes are available on the managed service and in your own cloud.

`/recall` returns ranked documents with source excerpts under a token budget. The application decides whether those documents establish an answer.

## Architecture differences

|  | Zep | past.dev |
| --- | --- | --- |
| Open-source engine | Graphiti, Apache 2.0 | No open-source engine |
| Storage underneath | Neo4j, FalkorDB or Amazon Neptune | Not part of the public API contract |
| Time input | Bi-temporal, facts invalidated rather than deleted | Occurrence timestamp and optional recall timestamp |
| What you send | Episodes: messages, text or JSON | Text plus the timestamp it happened, backdatable |
| Read surface | Retrieval of facts and context | `/recall`: ranked documents |
| Abstention | Your code decides | Your code evaluates returned documents |
| Cost visibility | Plan-level | Accounted per call on the console dashboard |
| Self-host | Run Graphiti yourself, with your own graph database | Your own cloud, on an Enterprise agreement |

## How to choose

Choose based on the requirements below.

- **Choose Zep or Graphiti** if you want the engine itself under an open-source licence, if you are already running a graph database, or if you want to write the traversals yourself.
- **Choose past.dev** if you want a memory API with timestamped ingest, identities, audiences, and source-backed recall that runs as a managed service or, on an Enterprise agreement, in your own cloud.
- **Both products** record time on facts. [Knowledge graphs for LLM applications](/knowledge-graph-for-llm) explains temporal graphs.

## What this page does not claim

This page contains no head-to-head scores. Our evaluation uses BEAM and LoCoMo with community-corrected answer keys, with full-context and plain-RAG baselines and several judges. The method is published on [benchmarks](/benchmarks). We will publish product comparisons after documenting a reproducible configuration for the other product.

## Related

- [Knowledge graphs for LLM applications](https://past.dev/knowledge-graph-for-llm)
- [What is agent memory?](https://past.dev/what-is-agent-memory)
- [past.dev vs mem0](https://past.dev/vs/mem0)
- [How the pipeline works](https://past.dev/docs/memory-api/how-it-works)
- [How we measure memory](https://past.dev/benchmarks)