Self-organising long-term memory substrate for agentic LLM workflows, grounded in Event Segmentation Theory (EST) and Predictive Processing (PP). Ingests multi-turn conversations, segments them into topically coherent episodes via LLM-powered boundary detection, distils durable semantic knowledge from each episode, and exposes a unified search surface for downstream reasoning. Designed as a minimalist production-ready core: PostgreSQL for structured metadata, Qdrant for vector similarity.
- Storage
- Two complementary stores: PostgreSQL holds episode metadata, temporal anchors, and structured knowledge records; Qdrant stores dense vector embeddings of episodes and knowledge facts for semantic search. Episodes are generated by an LLM-powered segmentation pipeline that applies transitional masking heuristics (from EST) to detect coherent topic boundaries before writing.
- Retrieval
- Unified search surface over both stores: semantic vector similarity via Qdrant for fuzzy recall, and structured PostgreSQL queries for exact metadata filters (time, source, entity). A Predict–Calibrate cycle extracts high-value facts from gaps between existing knowledge and new episodes, progressively refining the knowledge base. Async concurrency and caching keep latency low under load.
- Self-host
- Self-host: moderate
- License
- MIT
- Pricing
- Open-source MIT, free to self-host. Requires a running PostgreSQL 16 instance and a Qdrant instance (Docker Compose setup provided). Also requires an LLM and embedding API (OpenRouter or OpenAI-compatible). · Free / OSS
- GitHub stars
- 221
- Last release
- 2025-11-05
- Last commit
- 2026-04-16
- First catalogued
- 2026-06-28
Strengths
- Cognitive-science grounding (Event Segmentation Theory + Predictive Processing) produces semantically coherent episode boundaries — not arbitrary chunk splits
- Predict–Calibrate cycle actively surfaces memory gaps and fills them, making the knowledge base self-correcting over time
- Production-ready concurrency design (async, caching) and clean separation of metadata (Postgres) from vectors (Qdrant)
- MIT-licensed with a research paper (arXiv:2508.03341) providing theoretical foundations
Watch out
- Requires two external services (PostgreSQL + Qdrant) — heavier infrastructure footprint than single-binary memory solutions
- 204 stars, no formal release tag; actively developed but not yet at a stable versioned release — API may change
- LLM used for episode boundary detection adds cost and latency at write time; offline/low-cost paths are not documented
- Escalated needs-review → deprecated on 2026-11-01: the 90-day needs-review grace window (opened 2026-08-01) has elapsed with the repo still dormant (last commit 2026-04-16, ~199 days) and no maintainer revival — the same three community PRs (#18–#20) remain unmerged.
Best for
- Agentic LLM workflows needing structured long-term memory with semantically coherent episodes and a unified search surface across episodic and semantic stores
Benchmark results
No sourced results yet.
Sources
- Open pull requests — three community PRs unmerged (2026-05-29 uv lock refresh, 2026-06-03 semantic-memory dedup fix, 2026-07-22 evaluation dashboard), against a default branch last committed 2026-04-16; the basis for moving status active → needs-review (third-party)
- Monthly card re-verification — GitHub API repo metadata: 208 stars, license MIT confirmed, archived=false, last commit on the default branch 2026-04-16 (third-party)
- Nemori README (vendor)
- Nemori paper (arXiv) (paper)
- GitHub API repo metadata (204 stars, MIT, no formal release) (third-party)
- GitHub license API — confirms MIT; repo not archived, still 0 formal releases, last push 2026-04-16 (third-party)
- Nemori README (re-checked) — architecture, Postgres+Qdrant storage, and 'production-ready concurrency' framing unchanged; still no MCP/REST/CLI surface, install is pip install -e . from source (vendor)
- Monthly card re-verification — GitHub API repo metadata: 212 stars, latest tag v0.1.1 (2025-11-05); not archived. (third-party)
- Status lifecycle escalation — needs-review (since 2026-08-01, 92 days in review ≥ 90-day grace) escalated to deprecated: last commit still 2026-04-16 (199 days, ≥180-day dormancy threshold), not archived, PRs #18–#20 (opened 2026-05-29 to 2026-07-22) remain open/unmerged. GitHub metadata re-confirmed: 217 stars, MIT, not archived. (third-party)
Last verified 2026-11-01 · updated by refresh-framework-cards