open-index vs Memary
Agent memory and knowledge graph projects compared
open-index and Memary both add persistent memory and graph-based context to agents. open-index centers on schema-defined domain knowledge with SQLite or OpenSearch, while Memary emphasizes personalized memory across user interactions and graph retrieval.
open-index: Build Agent Memory on SQLite and OpenSearch
Open Index is a domain-agnostic memory layer for agents, built around a searchable context graph with SQLite and FTS5 by default, plus an OpenSearch backend. Define structured concepts and relationships, then let agents search and maintain that knowledge through MCP.
Memary: Add Memory to Autonomous Agents with LlamaIndex
Memary is an open-source memory layer for autonomous agents, built around LlamaIndex, knowledge graphs, and a ReAct chat agent. It stores interaction-derived entities and user context to help agents retrieve relevant information and tailor responses.
| open-index | Memary | |
|---|---|---|
| Language | Python | Jupyter Notebook |
| License | MIT | MIT |
| Stars | 112 | 2.7k |
| Forks | 5 | 205 |
| Last analyzed | Oct 21, 2026 | Oct 25, 2026 |
Key differences
- open-index lets teams define concepts, fields, entities, and relationships with YAML schemas; Memary tracks encountered entities, their relationships, frequency, and recency.
- open-index is written in Python and defaults to SQLite with FTS5, with OpenSearch for shared deployments; Memary is listed as Jupyter Notebook and offers FalkorDB or Neo4j, as well as local JSON storage.
- open-index provides a CLI, explorer UI, and MCP server for inspecting and maintaining structured knowledge; Memary includes a ReAct chat agent and tools for search, vision, location, and stock queries.
- open-index targets domain-specific knowledge for areas such as support, infrastructure, and sales; Memary focuses on personalized agent responses and distinct user contexts.
- open-index has 112 stars and 5 forks and is described as less than six months old; Memary has 2.7k stars and 206 forks.
- Both projects use the MIT license. open-index requires Python 3.10 or newer, while Memary specifies Python versions up to 3.11.9.
Choose open-index if you…
- need schema-defined, domain-specific knowledge that agents can read and update through MCP.
- want SQLite and FTS5 as a local default, with OpenSearch available for shared, multi-writer deployments.
- need CLI and UI tools to inspect entities, relationships, search activity, and connector jobs.
Choose Memary if you…
- want an agent memory layer that tracks entities across user interactions, including frequency and recency.
- need graph retrieval with recursive or multi-hop context assembly and external-search fallback.
- want to extend an included ReAct agent with tools and options for separate user contexts.
Compare the numbers side by side, or add more projects
This comparison is generated with AI from the OSRepos analyses of both projects. Always check each project's repository and documentation before choosing.