Home / Directory / Agents & orchestration / AI agent memory / Zep
Zep
Context platform for AI agents from Y Combinator's Winter 2024 batch that builds temporal knowledge graphs from conversations, documents, and business data.
Teams building customer-facing or enterprise agents that need context from chats, documents, and business systems, with access policies and audit logs.
Hobby projects that want a cheap paid tier, or teams that only need simple chat history without a knowledge graph.
Verdict
Zep is worth using for teams whose agents need to know how a customer or account has changed over time, not only what was said last. Its temporal knowledge graph tracks how facts change over time, which helps agents that work with the same accounts for months. Its governance features suit enterprise rollouts.
Score Breakdown
How Zep scores in the categories that matter to its buyers.
Pricing
Zep publishes credit-based self-serve prices. Annual billing saves 17%, and retrieval, storage, and users are not metered.
| Plan | Price | What stands out |
|---|---|---|
| Free | $0 | 10,000 credits a month; limits apply |
| Flex | $125 / month | 50,000 credits, then $25 per 10,000; 5 projects, 600 requests per minute |
| Flex Plus | $375 / month | 200,000 credits, then $75 per 40,000; webhooks, analytics, 10 projects |
| Enterprise | Custom | SOC 2 Type II and HIPAA BAA, audit logs, own-key and in-VPC deployment |
The Field at a Glance
How Zep compares with other AI agent memory vendors we reviewed, by Overall Score and relative typical engagement cost.
Zep scores 7.9 in AI agent memory, under Mem0 (8.0), and ahead of Letta (7.8), Supermemory (7.8), and Cognee (7.7). Relative cost lands upper-mid for this subcategory.
Use-case matrix
| Use case | Fit | Notes |
|---|---|---|
| Facts that change over time | Strong | Facts are dated and old ones are invalidated. |
| Source tracing | Strong | Each fact links back to the record it came from. |
| Access policies | Strong | Policies control what context each agent can read. |
| Open-source path | Strong | Graphiti is the open-source graph framework behind Zep. |
| Low-cost paid tier | Weak | Paid plans start at $125 a month. |
| Compliance on self-serve | Weak | SOC 2 report and HIPAA BAA are listed on Enterprise. |
Who it’s for
Good fit
- Enterprise agent teams
- B2C apps with long customer histories
- Teams needing audit trails for agent context
Poor fit
- Hobby projects
- Simple chat history needs
- Buyers wanting compliance on a self-serve plan
Review Excerpts
Below are excerpts from public reviews. Our team scoured public reviews, forums, and chat rooms to get a balanced view of customers' experience with this company. Paid reviews and pay-for-play sites such as Clutch were excluded.
“Zep AI was instrumental in enabling the Sidekick's personalized experience through dynamic memory retrieval.”
“Unlike other systems that only retrieve static documents, Zep uses a temporal knowledge graph to combine conversations and structured business data, keeping track of how things change over time.”
Zep's plan comparison lists its SOC 2 Type II report and HIPAA BAA only on Enterprise, and self-serve plans start at $125 a month after the free tier.
Company Facts
- Founded
- 2023 (Y Combinator)
- Headquarters
- San Francisco (Y Combinator)
- Employees
- About 10 (Y Combinator)
- Accelerator
- Y Combinator, Winter 2024 batch (Y Combinator)
- Website
- getzep.com
Key Leadership
Daniel Chalef
Founder and CEO
Methodology
This page is an independent evaluation of Zep for buyers comparing options in AI agent memory. AI Industry Reviews accepts no sponsorships, advertising, or pay-for-placement fees. Zep did not pay for this review.
What we scored
The headline number is an Overall Score on a 0-10 scale. Eight criteria sit under it in two groups.
- Buyer outcomes (for AI agent memory)
- Temporal context graph
- Retrieval speed and accuracy
- Enterprise governance
- Self-serve entry price
- Company & commercial
- Innovation & product leadership
- Project management & communication
- Pricing
- Contract fairness
Pricing measures whether the price looks fair for the value delivered, including packaging and renewal friction that show up in real buying cycles.
Score Composition
| Input | Weight | What it covers |
|---|---|---|
| Reviews | 40% | A proprietary read of what practitioners say about likes, complaints, and day-to-day use, including public review sites, forums, and private chat rooms. Paid reviews and pay-for-play sites such as Clutch are out of scope. |
| Product | 35% | Hands-on look at screens and workflows. |
| Pricing | 15% | Whether the price looks fair for what you get. |
| Docs & training | 10% | Docs, tutorials, and training material. |
How we balanced the evidence
The Overall Score is the simple average of the eight criteria. Recommendation language follows that score and the fit pattern described above.
Scope
Zep is graded here as AI agent memory. Criteria scores can move as more review volume and product checks are added.