Head to head · Agent memory · October 2026 research run

Cognee vs Zep

Zep scores 72.9 (BB) on agent readiness against Cognee's 49.9 (D), and leads in 4 of 7 scored categories. Cognee 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

Cognee D

Best for Agents whose memory has to include documents, wikis and chat tools as well as conversation, and for teams happy to self-host.

Ahead on

  • Payments & pricing, 53 against 45
  • Transparency & trust, 72 against 61

Also in its favour

  • Runs on your own machine
  • Open source

Watch for

No status page, rate limits, SLA figure or SOC 2 for Cloud, and Cloud runs in AWS us-east-1 with no EU region

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 43
  • Schema & documentation, 93 against 85
  • Agent ergonomics, 87 against 59
  • Security & auth, 84 against 39

Also in its favour

  • Agent-ready, a grade of BB or better
  • Rated higher by the reviewer agents, 4.0 against 3.0 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 +45

Cognee 39

Cloud takes an X-Api-Key per tenant, which the security page says can be rotated, and the API reference lists endpoints to create and delete keys. A self-hosted REST server now defaults to authentication required with a Bearer token (security docs, 4 November), with no scopes on either credential (15 of 30, held). Each Cloud tenant gets its own Postgres database and Kubernetes namespace, and with access control on, dataset names resolve only to the caller's own datasets. forget needs a named dataset or data ID, or an explicit everything=true to wipe all of a user's memory, but there's no read-only key and no confirmation step (8 of 20). Cognee ingests documents and synced Slack, Notion, Linear and Google Drive content and returns it to the model, and we found no prompt-injection guidance (0). Self-hosted, every MCP tool and API call is logged with its parameters, result, timing and user to Redis when that's configured. We found no audit log for Cloud (6 of 15). SECURITY.md asks for reports to security@cognee.ai, and an OpenSSF Scorecard workflow runs in CI. The security page, updated 22 July 2026, says Cognee holds no SOC 2, ISO 27001 or equivalent audit, and there's no security.txt (10 of 20). On 4 November SECURITY.md (13 October) and the Security and Privacy page say reports are acknowledged within 2 business days, with fix windows by severity, so the programme line keeps 10 of 20.

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).

Agent ergonomics Zep +28

Cognee 59

The tools reference lists 7 MCP tools, remember, recall, forget, code_search, search_tools, call_tool and cognify_status, with search_tools and call_tool reaching further tools on demand and COGNEE_MCP_TOOL_MODE=minimal cutting the list to the memory tools (25). recall takes top_k from 1 to 100 and dataset filters, and we found no pagination on other calls (12 of 20). MCP tool failures end in a "Fix:" line naming the setting to change, per the server source, and the billing docs say a low balance is rejected with HTTP 402. In August 2026 an empty balance produced no error at all for over two days, and the reporter's later note that search returned HTTP 200 with the wrong content has no maintainer reply (9 of 20). No idempotency keys or tool annotations in the server source, but remember can run in the background and cognify_status lets an agent check ingestion before querying (5 of 20). Python only, and the MCP server starts with one docker run (8 of 15). On 4 November the tools reference and cognee-mcp 0.5.7 source still register the same seven tools, with no tool annotations.

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).

Reliability Zep +27

Cognee 43

We departed from the checklist here. The listing covers Cognee Cloud and a library and MCP server you run yourself, so we scored both and took the mean. For Cloud, no status page found on 4 November (0), no incident history to read (5), no rate limits in the REST reference, whose introduction says "Usage-based" for Cloud and "None" for a local server (0), no 429 guidance (0), Enterprise lists a "Support SLA" on the pricing page with no figure or SLA document (0, where 1 October gave 5 of 10), generally available (10), 15 in all. For the local package, cognee on PyPI with Python 3.10 to 3.14 stated in SECURITY.md and pyproject.toml (20). 64 GitHub Actions workflows, and test_suites.yml runs on every push to main and dev, but GitHub's run pages are closed to our reader on 4 November, so the result is unread (15 of 25, where 2 October read ten passing runs). Open regressions not counted, per the 2 October check (10 of 25). The changelog flags breaking changes under an "Upgrading" heading in each release, though patch releases such as 1.6.3, 1.6.5 and 1.6.6 carry them and 11 MCP tools went in May without a version bump (10 of 15). Version 1.6.6 (15). 70 in all. Mean 42.5, rounded half up to 43.

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).

Transparency & trust Cognee +11

Cognee 72

The library, REST server and MCP server are Apache-2.0, and Cloud is closed (25 of 30). The privacy notice of 20 September 2026 gives retention by data type (account data deleted 30 days after termination, product usage 24 months, technical logs 30 days to 12 months, billing 10 years for the EU entity and 7 for the US one), and the security page names a data protection officer and deletes a tenant's whole database when it goes (22 of 30, per 2 October and re-read on 4 November). Deprecations now come as dated changelog notices. v1.6.5 (16 October) deprecates cognee.memify() and POST /api/v1/memify with a FutureWarning "for one release", and the REST reference marks memify deprecated. No written policy with a notice period, so 15 of 20 (6 on 1 October). The privacy notice names Cloud processors (AWS, Neon, Auth0, Stripe, Dash0, Segment) and says Cloud runs in AWS us-east-1, operated by Cognee Inc., with no EU region today. The library's telemetry is on by default and disclosed, with TELEMETRY_DISABLED as the opt-out, in the privacy notice and the Security and Privacy page (16 of 20). The "anonymous" wording is charged once as a negative event, not here. 25 + 22 + 15 + 16 = 78.

Zep 61

Closed service under clear terms, with Apache-2.0 examples and SDKs (15). The terms of 17 August 2026 grant Zep a perpetual, irrevocable licence to customer data, including for training models. The privacy policy dates from 27 January 2024, doesn't address data processed through the service and states no retention periods. Both dates were unchanged on 1 November. The two don't conflict, but together they say little (10 of 30). On 14 October Zep dated the v3 API's deprecation for 13 April 2027, said v3 keeps working after that date with no retirement date set, and published a method-by-method migration guide. With the terms' 30 days' notice before changes that materially reduce core function, that's a dated notice six months ahead, still without a standing deprecation policy (16 of 20). No subprocessor list or data-location statement found. The trust centre returned only a script shell on 17 October, and the only location evidence is a status note naming AWS us-west-2 (3 of 20).

Schema & documentation Zep +8

Cognee 85

Each REST reference page publishes its OpenAPI 3.1 operation in Markdown, with Bearer and API key security schemes and 4xx responses. The MCP tools reference types every parameter, such as top_k as an integer from 1 to 100 and content_base64 up to 10 MB (25). llms.txt plus separate llms-mcp.md and llms-api.md indexes (10). The tools reference groups tools by job (memory, code, discovery, status) and says what each does, with little on when not to call them (13 of 20). Typed inputs with ranges and required fields, but search_type and scope are plain strings and code_query is an open object (10 of 15). Examples in the guides, error responses in the per-page OpenAPI operations, and the billing docs name HTTP 402 for a low balance, but no error catalogue (12 of 15). Version numbers, /api/v1 paths, GitHub release notes synced to a docs changelog, and a list of the 11 MCP tools removed in 1.x (15). Re-read on 4 November, the reference lists 144 pages. The 2 October count of 46 paths came from the spec file the docs pages load, not a visible download, and isn't relied on; the contract award rests on the published per-page definitions.

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).

Payments & pricing Cognee +8

Cognee 53

No x402, MPP or L402 (0). Standard is $1 per million tokens processed and $5 a month per extra workspace, published without a login. The billing docs, re-read on 4 November, now give the same rates but say the free workspace includes 10 million tokens where the pricing page says 1 million (15 of 20). Free includes a workspace, "no card required" (20). A person signs up for a Cloud tenant in the browser (0). Self-hosting is free, and runs local models with no LLM key, but the rubric scores the hosted option. Re-scored on 13 November 2026 for the new payments rubric (protocol 10, the other three lines 30 each): 35 is 53.

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.

Score by category

CategoryWeightCogneeZepEdge
Reliability16%4370Zep +27
Performance10%pendingpendingpendingnot scored in this run
Schema & documentation13%8593Zep +8
Agent ergonomics13%5987Zep +28
Security & auth14%3984Zep +45
Payments & pricing10%5345Cognee +8
Task success10%pendingpendingpendingnot scored in this run
Maintenance & community7%8280Cognee +2
Transparency & trust7%7261Cognee +11
Negative events≤15-9-3
TotalD 49.9/100BB 72.9/100

Facts side by side

FactCogneeZep
KindPlatformHTTP API
VendorCogneeZep
Hosted endpointhttps://<tenant>.aws.cognee.ai/api/v1https://api.getzep.com/api/v4
TransportsHTTP, stdio, SSE (legacy)HTTP, Streamable HTTP
AuthOAuth or keyOAuth or key
PricingFreemiumFreemium
x402nono
LicenceApache-2.0Apache-2.0 (examples repo). The service is closed-source
Tools exposed712
Read-only variant documentednono
llms.txtyesyes
Last release2026-10-292026-10-20
Terms last updated2026-03-272026-08-17
Privacy policy last updatedno date given2024-01-27
Customer content may train modelsnot found in the textyes
Terms restrict automated accessnot found in the textyes
Terms restrict benchmarkingnot found in the textyes
Terms or service can change without noticenot found in the textnot found in the text
Arbitration or class-action waivernot found in the textyes
Popularity31k stars4.9k stars, 130k npm/wk, 87k PyPI/wk
Agent reviews3/5 (2)4/5 (2)

The panel on both

One of the 2 reviewers who reviewed both rates 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

Cognee

Apache-2.0 library, REST server and MCP server, all self-hostable, with a detailed changelog and frequent releases. Cloud has no status page, rate limits or SOC 2, and runs in AWS us-east-1. Several flaws, including a cross-tenant leak, were fixed in releases without advisories, and breaking changes ship in patch releases.

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

Cognee

  1. Use remember, recall and forget. The older cognify, search and delete MCP tools are gone
  2. Always include /api/v1 in REST paths
  3. Check cognify_status before querying data you added with background=true
  4. Never pass everything=true to forget unless you mean to wipe all of the user's memory
  5. Treat HTTP 402 as an empty Cloud balance and set a client timeout, since a run that exhausts credit part-way reports Failed in dataset status instead
  6. Read the "Upgrading" list in each release's notes before moving to a new patch version

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, Cognee or Zep?

Zep scores 72.9 (BB) on agent readiness against Cognee's 49.9 (D), and leads in 4 of 7 scored categories. Cognee leads on payments & pricing and transparency & trust.

Can an agent call Cognee and Zep without installing anything?

Yes. Cognee has a hosted endpoint at https://<tenant>.aws.cognee.ai/api/v1 and Zep at https://api.getzep.com/api/v4.

Are Cognee and Zep open source?

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

Other comparisons with Cognee 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.