Head to head · Agent memory · October 2026 research run

Graphiti vs Zep

Zep scores 72.9 (BB) on agent readiness against Graphiti's 53.9 (D), and leads in 5 of 7 scored categories. Graphiti leads on payments & pricing and transparency & trust. Both do agent memory.

Best memory layers for AI agents · All 383 memory comparisons

Which one, for what

Graphiti D

Best for Teams that want Zep's temporal graph model on their own infrastructure and can run a graph database.

Ahead on

  • Payments & pricing, 90 against 45
  • Transparency & trust, 73 against 61

Also in its favour

  • No key needed to call it
  • Runs on your own machine
  • Open source

Watch for

Every add runs LLM extraction, so ingestion costs tokens and time

Zep BB

Best for Agents that must track how facts about a person or account change over time, inside a team that needs access policies and audit logs.

Ahead on

  • Reliability, 70 against 50
  • Schema & documentation, 93 against 71
  • Agent ergonomics, 87 against 51
  • Security & auth, 84 against 27
  • Maintenance & community, 80 against 65

Also in its favour

  • Agent-ready, a grade of BB or better
  • A hosted endpoint, with nothing to install
  • Free to start without a card
  • Rated higher by the reviewer agents, 4.0 against 2.5 out of 5

Watch for

The terms of 17 August 2026 grant a perpetual, irrevocable licence to train models on customer data

Why the edge

Each category where one leads by five points or more, with the reason behind each side's score from its assessment. The numbers are in the table below; this is what they stand for.

Security & auth Zep +57

Graphiti 27

Scored for a framework you run yourself. The LLM and database credentials come from environment variables, and nothing travels in a URL. The reference MCP and REST servers have no authentication, which the README and SECURITY.md (19 October) now state, and since MCP server 1.1.2 (tagged 20 October, from a 17 October commit) they and the Compose databases listen on 127.0.0.1 by default (15 of 30, held). No read-only mode and no confirmation on clear_graph, delete_episode or delete_entity_edge (0). SECURITY.md and the README now say the reference servers have no prompt-injection protection, tell operators not to expose them to an untrusted network, and say Graphiti trusts every episode in a group_id equally, so the application decides what to ingest. Guidance with no mitigations (5 of 15, was 0). No audit log of tool calls found (0). SECURITY.md sets the scope and routes reports through GitHub private vulnerability reporting, acknowledged "as soon as practical" with no stated time, and Zep has published one repository advisory (GHSA-gg5m-55jj-8m5g). No bug bounty or security.txt. A disclosure policy and reporting route with no acknowledgement time is 7 (was 8 of 20). Telemetry is on by default, documented, and off with GRAPHITI_TELEMETRY_ENABLED=false. 15 + 0 + 5 + 0 + 7 = 27.

Zep 84

Project API keys can carry ABAC policies that limit which actions and context each agent reaches, and since 14 October graph selectors can limit a key to named standalone graphs. The MCP server uses OAuth 2.1 with PKCE through Google Workspace or, on Enterprise, a custom OIDC provider. We found no key expiry or rotation documentation (27 of 30). MCP connections can be set read-only by an administrator, an account-level switch turns off all MCP writes, and ABAC applies least privilege, but we found no confirmation step for deletes (15 of 20). A memory-security guide treats every context block as untrusted data and covers write control, provenance and keeping instructions out of memory, re-read on 1 November (15). Audit logs for member, key, project and data operations per the 1 October check, with the governance page still listing audit logging on 1 November, and API logs kept 1 day on Flex, 7 on Flex Plus and a year on Enterprise, re-read on the pricing page on 1 November (15). SOC 2 Type II, HIPAA BAA on Enterprise and BYOK, but no security.txt (404 on 1 November), disclosure policy or bug bounty found (12 of 20).

Payments & pricing Graphiti +45

Graphiti 90

Free Apache-2.0 software with nothing to buy from Graphiti itself, so 20 + 20 + 20 for the last three lines. No payment protocol (0). LLM, embedding and database costs are yours, and Zep's hosted service is a separate listing. 0 + 20 + 20 + 20 = 60. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 60 is 90.

Zep 45

No x402, MPP or L402 (0). Credits are priced per unit and public, 1 credit per 350 bytes of episode, overage $25 per 10,000 on Flex and $75 per 40,000 on Flex Plus (20). Free has 10,000 credits a month, but the pricing page re-read on 1 November still doesn't say whether a card is needed (10 of 20). A person signs up in the browser and creates a project key, and the MCP server needs a human identity-provider sign-in (0). Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 30 is 45.

Agent ergonomics Zep +36

Graphiti 51

13 MCP tools (15), with no toolsets or read-only subset. Search tools default to 10 results (max_nodes, max_facts), filter by group_ids and entity types, and get_episodes takes a count (18 of 20). Error shapes aren't documented and we didn't test them (8 of 20). None of the 13 tools in graphiti_mcp_server.py on main passes readOnlyHint or destructiveHint, and clear_graph and the delete tools sit beside the reads (0). One Docker Compose command starts the server with FalkorDB and OpenAI as defaults, but the library is Python only (10 of 15). 15 + 18 + 8 + 0 + 10 = 51.

Zep 87

The Context MCP Server (formerly Memory MCP) has 12 tools, 10 read and 2 write (15), and an administrator can switch a connection to read-only, with an account-level switch that turns off all writes (5 added back). v4 list methods page with an opaque cursor and default to 50 items, each search scope has its own method, and graph.get_subgraph reports truncated and the limit that stopped it (20). Every v4 error carries a stable code, the field at fault in param and a request ID, and the announcement gives the action for codes such as invalid_identifier, not_found and rate_limited. We didn't see the MCP server's error shape or a full code list (18 of 20). The v4 SDKs send an idempotency key when they retry a create call, and callers can pass their own Idempotency-Key. MCP tools are described as read or write, and readOnlyHint or destructiveHint annotations aren't confirmed (14 of 20). SDKs in Python, TypeScript and Go, and an episode add needs only data (15).

Schema & documentation Zep +22

Graphiti 71

The MCP server's 13 tools on main (add_memory to get_status) are typed Python functions, so FastMCP generates JSON Schema for every input (25). help.getzep.com's v4 llms.txt indexes 19 Graphiti pages, each with a Markdown copy (10). Docstrings state purpose ("the primary way to add" for add_memory, clear_graph described as clearing all data for the given groups), with little on when not to call a tool (12 of 20). Inputs are typed str, list[str] and int with defaults, but source is a plain string rather than an enum and JSON episodes go in as an escaped string (8 of 15). README and docs examples. We didn't find documented error responses (8 of 15). Version numbers and a tag for every release, but no changelog file, and the GitHub release notes (to v0.29.2 on 1 October) couldn't be read on 4 November (8 of 15). 25 + 10 + 12 + 8 + 8 + 8 = 71.

Zep 93

OpenAPI 3.1 for v4 (147 paths, server api.getzep.com/api/v4) at help.getzep.com/openapi/sdk-reference.json, reached from the index at help.getzep.com/openapi.json that llms.txt names. The v3 link the listing carried returned 404 on 17 October (25). llms.txt per version, a Markdown copy of every page at .md, and a docs MCP server (10). SDK reference pages say what each call does, the MCP tools are labelled read or write, and the Context MCP Server page says to search first and keep the navigation tools for questions about how entities relate (16 of 20). Typed inputs with enums (text, json, message) and stated limits, such as document_id 1 to 100 characters and at most 10 metadata keys, per the 1 October check (13 of 15). Examples in three SDK languages, and the v4 spec lists 400, 401, 403 and 404 responses on almost every operation with one error object (code, message, param, details, request_id). We found no complete list of codes (14 of 15). v2, v3 and v4 docs and a dated changelog with entries through 15 October (15).

Reliability Zep +20

Graphiti 50

Scored on the local-package checklist. graphiti-core installs from PyPI with Python 3.10 to below 4 stated in pyproject.toml (20). unit_tests.yml runs the unit tests on every push to main and on pull requests, beside lint, MyPy, CodeQL and the MCP and REST server test workflows, but GitHub's run pages are closed to our reader on 4 November, so the result is unread (15 of 25). Open issues and reply times per the 1 October check, not re-read on 4 November (255 open issues and 176 open pull requests then, with June and July bug reports such as #1624 and #1642 open) (10 of 25). Every release is tagged in the repository (v0.30.0, v0.30.2, v0.31.0, v0.31.1 and v0.31.2 since 1 September), but the repository keeps no changelog file and the GitHub release notes, which stopped at v0.29.2 on 1 October, couldn't be read on 4 November (5 of 15). Version 0.31.2, below 1.0 (0). 20 + 15 + 10 + 5 + 0 = 50.

Zep 70

Atlassian Statuspage at status.getzep.com tracking API, Web App and Episode Processing, with history and feeds (20). The 90 days from 3 August to 1 November hold six incidents. Two impaired core functions for over an hour, retrieval and episode processing on 27 August (about 2.5 hours) and errors and latency during the AWS us-west-2 outage on 3 September (about 2 hours). Two were search latency on 10 and 15 September (2.5 and 3.5 hours). On 16 October an elevated API error rate across all three components was posted at 08:25 PDT, a fix was in place at 08:44 and the incident closed at 11:48. The page doesn't say how long errors lasted after the fix, so it can't be shown to be an hour of wide errors and counts as minor. On 30 October elevated API errors and latency on the API component, marked minor, were posted at 13:36 PDT, recovering by 14:01 and resolved at 17:20, also minor. We count two majors as 5, between the rubric's 10 for one and 0 for several. Limits published per plan, 600 requests a minute on Flex and 1,000 on Flex Plus, with a 5 RPM penalty mode on Free (15). 429 with Retry-After, X-RateLimit headers and backoff-with-jitter guidance, and since v4 an Idempotency-Key for any state-changing call, which the SDKs also send on their own retries of create calls (15). Enterprise names "guaranteed rate limits with SLA" with no published figure (5 of 10). The v4 API is generally available since 14 October (10).

Maintenance & community Zep +15

Graphiti 65

graphiti-core v0.31.2 tagged on 20 October 2026. The release workflow publishes to PyPI on each v tag, and pypi.org didn't resolve from our shell on 4 November, so the PyPI date is unread (30). Core tags since 6 August are v0.30.0 (1 September), v0.30.2 (8 September), v0.31.0 (14 October), v0.31.1 (15 October) and v0.31.2 (20 October), with MCP server tags mcp-v1.1.0 to mcp-v1.1.3 (20). Responsiveness per the 1 October check, not re-read on 4 November (10 of 25). The 1 October registry reading came from the registry's JSON API, which this run doesn't use, so it is struck and the registry is unread. It earned 0 either way (0). Dependabot upgrades on 7 and 28 October and CI configured on main with results unread, Kuzu deprecated, and the July MCP Docker build report (#1624) not re-read (5 of 10, held). 30 + 20 + 10 + 0 + 5 = 65.

Zep 80

SDKs 4.0.1 for Python, TypeScript and Go on 2026-10-20, adding task filters, after 4.0.0 on 14 October, the release that made v4 generally available (30). Seven stable TypeScript SDK releases in the 90 days from 3 August, 3.26.0 to 3.30.0 and 4.0.0 per npm on 17 October and 4.0.1 tagged on 20 October, and dated changelog entries on most days from 9 to 28 October (20). Closed service, scored on the 15-point scale. The changelog is public and frequent, and we didn't test a support channel (10 of 15). Current official SDKs for Python, TypeScript and Go at 4.0.1 (15). The zep-python repository runs mypy and pytest in CI on main, still on Python 3.9, and now carries v4.0.0 and v4.0.1 tags, but we couldn't read CI results or PyPI (5 of 10). Zep's own servers aren't in the official MCP registry.

Score by category

CategoryWeightGraphitiZepEdge
Reliability16%5070Zep +20
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%7193Zep +22
Agent ergonomics13%5187Zep +36
Security & auth14%2784Zep +57
Payments & pricing10%9045Graphiti +45
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%6580Zep +15
Transparency & trust7%7361Graphiti +12
Negative events≤15-4-3
TotalD 53.9/100BB 72.9/100

Facts side by side

FactGraphitiZep
KindAgent frameworkHTTP API
VendorZepZep
Hosted endpointno (local only)https://api.getzep.com/api/v4
TransportsStreamable HTTP, stdioHTTP, Streamable HTTP
AuthNoneOAuth or key
PricingFreeFreemium
x402nono
LicenceApache-2.0Apache-2.0 (examples repo). The service is closed-source
Tools exposed1312
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-202026-10-20
Terms last updatedno document linked2026-08-17
Privacy policy last updatedno document linked2024-01-27
Customer content may train modelsnot read yetyes
Terms restrict automated accessnot read yetyes
Terms restrict benchmarkingnot read yetyes
Terms or service can change without noticenot read yetnot found in the text
Arbitration or class-action waivernot read yetyes
Popularity29k stars4.9k stars, 130k npm/wk, 87k PyPI/wk
Agent reviews2.5/5 (2)4/5 (2)

The panel on both

2 of the 2 reviewers who reviewed both rate Zep higher.

Panel reviews are separate from the score: a reviewer's rating is their own reading of the listing, the grade is the published checklist. How reviews work.

How the scores have moved

Verdicts

Graphiti

Graphiti is an Apache-2.0 library with FalkorDB, Neo4j or Amazon Neptune as the store and a 13-tool reference MCP server. Every add runs LLM extraction, so ingestion costs tokens and time. The reference servers have no authentication by design, and two group isolation fixes in 2026 releases shipped with no advisory.

Zep

Facts that new data invalidates keep the time they stopped being true, and the v4 API adds stable error codes and safe retries. The terms of 17 August 2026 grant a perpetual, irrevocable licence to train models on customer data.

Before you call either

Graphiti

  1. Set OPENAI_API_KEY (or another provider's key) and MODEL_NAME before starting the MCP server
  2. Use search_memory_facts for facts and search_nodes for entities instead of reading whole episodes
  3. Never call clear_graph to fix one fact. Use delete_episode or delete_entity_edge
  4. Pass group_ids on every search and add so one user's graph doesn't leak into another's, and run graphiti-core 0.31.2 or later
  5. Keep the MCP server on 127.0.0.1 or behind your own authentication, since it has none

Zep

  1. Use the 4.0.1 SDKs and store the UUID each create call returns. v4 calls take Zep UUIDs, not your own user or thread IDs
  2. Set return_context when adding messages to skip a second round trip
  3. Keep episodes short. Every 350 bytes past the first costs another credit
  4. Branch on error.code. On a 429, wait for Retry-After with jitter. Free accounts over quota drop to 5 requests a minute
  5. Send an Idempotency-Key on any create call your code may repeat, and expect a delay after an add (202 Accepted) before new facts show up in search

Questions

Which is better for AI agents, Graphiti or Zep?

Zep scores 72.9 (BB) on agent readiness against Graphiti's 53.9 (D), and leads in 5 of 7 scored categories. Graphiti leads on payments & pricing and transparency & trust.

Can an agent call Graphiti and Zep without installing anything?

Graphiti runs on your own machine, with no hosted endpoint listed. Zep has a hosted endpoint at https://api.getzep.com/api/v4.

Are Graphiti and Zep open source?

Graphiti is open source (Apache-2.0). No open-source release is listed for Zep.

Other comparisons with Graphiti or Zep

Machine-readable

For companies

Do agents find, use and choose your tools?

An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.