MemOS vs spector
MemTensor's memory OS for LLM agents: one API over graph-structured, multi-modal memory with hybrid retrieval and skill evolution — hosted, self-hosted (Neo4j + Qdrant) or local plugins. — versus — Java cognitive memory engine for agents: working, episodic, semantic and procedural tiers with decay, consolidation and association graphs, SIMD hybrid recall, served over MCP, REST/gRPC and SDKs.
Both pitch a memory 'OS' with typed memory and hybrid retrieval; MemOS has the research backing and broader adoption.
| MemOS | spector | |
|---|---|---|
| Stars | 12k | 28 |
| Forks | 1.1k | 16 |
| Language | TypeScript | Java |
| License | Apache-2.0 | Apache-2.0 |
| Last activity | today | yesterday |
| Topics | memory, skills | memory |
| Curated connections | 7 | 3 |
MemOS — the curator's take
Pick MemOS when you want a full memory stack rather than a vector wrapper: graph-structured memories you can inspect and correct in natural language, isolated or shared 'memory cubes' across users and agents, async ingestion, and traces that crystallize into reusable skills. The fastest path is a plugin — local SQLite for Hermes, OpenClaw or DeepSeek Harness, zero infra. Self-hosting the service means running Neo4j and Qdrant plus LLM and embedder keys, which is heavy for a single-user bot. Read the headline LoCoMo/LongMemEval numbers with care: the comparison runs on OmniMemEval, MemTensor's own harness. Skip it if you only need session recall for one coding agent — claude-mem or engrim is far less machinery.
spector — the curator's take
Consider it if you want memory that forgets and consolidates like a cognitive model rather than an ever-growing vector store, with an `npx` MCP server that embeds the kernel. Tiny project needing JDK 25 preview features — not something to bet production on yet; mature options like Hindsight or MemOS have far more evaluation behind them.