memorg Python External memory layer for LLM apps. Hierarchical SQLite + vector search, deterministic recall, MCP server.
Memorg is a pip-installable external memory layer for LLM apps: hierarchical SQLite storage, a USearch vector index, OpenAI embeddings and an MCP server. Retrieval blends semantic similarity, recency and importance within a token budget — no managed service.
Install
$ pip install memorgFrequently asked questions
What is memorg?
memorg is External memory layer for LLM apps. Hierarchical SQLite + vector search, deterministic recall, MCP server. It is part of the Neul Labs open-source portfolio (36 MIT-licensed projects for AI agent infrastructure). Memorg is a pip-installable external memory layer for LLM apps: hierarchical SQLite storage, a USearch vector index, OpenAI embeddings and an MCP server. Retrieval blends semantic similarity, recency and importance within a token budget — no managed service.
How do I install memorg?
Install with: `pip install memorg`. memorg is MIT licensed and ships prebuilt binaries where applicable.
What language is memorg written in?
memorg is written in Python.
Is memorg open source?
Yes. memorg is MIT licensed and publicly available at https://github.com/neul-labs/memorg. All 36 Neul Labs projects are MIT licensed.
How does memorg compare to alternatives?
See the detailed comparison at https://www.neullabs.com/compare/memorg-vs-mem0.