{"slug": "zep-gives-my-agent-perfect-recall-it-still-can-t-answer-average-per-group-about", "title": "Zep gives my agent perfect recall. It still can't answer \"average per group\" about its own memory.", "summary": "A developer compared Zep's temporal knowledge graph with nlqdb for agent memory, finding that while Zep excels at recall and entity resolution, it cannot perform aggregations like GROUP BY. nlqdb, which provisions a real Postgres database queried in English, handles analytical queries such as averages per group. The developer suggests using Zep for recall and nlqdb for analytical storage.", "body_md": "*Originally published at nlqdb.com/blog*\n\nIf you've wired up [Zep](https://www.getzep.com) you know the pitch: it's the Context Lake — a temporal knowledge graph (Graphiti, 27k+ stars) that stores every fact your agent learns as a node with a validity window, resolves entities, and hands back the most relevant facts at query time. For *recall* it's genuinely good, and it publishes benchmarks (LongMemEval, DMR) to prove it.\n\nBut we kept hitting the same wall. Once the agent had logged a few hundred things, we wanted to ask questions *about* the memory, not retrieve from it:\n\nA knowledge graph has no query planner. It returns relevant facts and hopes the LLM does the arithmetic — which is a hallucination generator, not a `GROUP BY`\n\n.\n\nThe honest split (the full side-by-side lives at [nlqdb vs Zep](https://dev.to/vs/zep/)): Zep wins on temporal validity, entity resolution, and vector recall over conversation. nlqdb wins when the agent needs to **aggregate** its memory — it's a real Postgres the agent provisions and queries in English, so `GROUP BY / JOIN / HAVING`\n\nactually work. They compose: Zep the recall layer, nlqdb the analytical store. Pick the one that matches the question you actually need answered.\n\n(Landscape facts verified 2026-06-19; both products' weaknesses are in the comparison, not just ours.)", "url": "https://wpnews.pro/news/zep-gives-my-agent-perfect-recall-it-still-can-t-answer-average-per-group-about", "canonical_source": "https://dev.to/omer_hochman/zep-gives-my-agent-perfect-recall-it-still-cant-answer-average-per-group-about-its-own-memory-3ae0", "published_at": "2026-07-19 07:28:37+00:00", "updated_at": "2026-07-19 08:27:48.468927+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "ai-infrastructure"], "entities": ["Zep", "nlqdb", "Graphiti"], "alternates": {"html": "https://wpnews.pro/news/zep-gives-my-agent-perfect-recall-it-still-can-t-answer-average-per-group-about", "markdown": "https://wpnews.pro/news/zep-gives-my-agent-perfect-recall-it-still-can-t-answer-average-per-group-about.md", "text": "https://wpnews.pro/news/zep-gives-my-agent-perfect-recall-it-still-can-t-answer-average-per-group-about.txt", "jsonld": "https://wpnews.pro/news/zep-gives-my-agent-perfect-recall-it-still-can-t-answer-average-per-group-about.jsonld"}}