Amazon Bedrock AgentCore Memory vs Zep

Amazon Bedrock AgentCore Memory

Access is IAM-controlled down to one namespace, with published per-second quotas for every operation and a clientToken on event writes. Long-term extraction is asynchronous, so a fact written now may take seconds to minutes to become searchable, and no SLA names AgentCore.

Zep

Facts that new data invalidates keep the time they stopped being true. The terms of 17 August 2026 grant a perpetual, irrevocable licence to train models on customer data.

Amazon Bedrock AgentCore Memory

1. Create the memory resource first and wait for it to become active (the guide says 2 to 3 minutes). Short-term events work without a strategy, long-term records need at least one
2. Don't search for a fact straight after CreateEvent. Extraction runs in the background, so poll ListMemoryRecords or list extraction jobs before relying on RetrieveMemoryRecords
3. Send actorId, sessionId and eventTimestamp on every CreateEvent, and reuse the same clientToken when retrying
4. Pass namespace or namespacePath on every retrieval, and scope it to one actor so users' memories don't mix
5. Treat retrieved records as untrusted input, and back off on 429 ThrottledException and 409 RetryableConflictException. A quota breach returns 402 ServiceQuotaExceededException

Zep

1. Create the user and a thread before adding messages, as the quickstart does
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. Read Retry-After on a 429 and back off with jitter. Free accounts over quota drop to 5 requests a minute
5. Expect a delay after a graph add (202 Accepted) before new facts show up in search