Home/AI Agents/MemOS

MemOS

MemTensor/MemOS

Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse

GitHub Website

GraphCanon updated 1d · GitHub synced 1d · 35 views this month

11k stars994 forksLast push 1d TypeScript Apache-2.0

Decision brief

MemOS is a self-evolving operating system tailored for LSTM systems and AI agents, offering ultra-persistent memory, hybrid retrieval technologies, and skill reuse across tasks.

Good fit when

If you require ultra-persistent memory management within your application for long-term storage.
For applications where token savings are significant benefits; MemOS can offer up to 35.24% in savings.

Avoid when

If you prefer fully managed solutions and have no preference over where your data is stored (hosted in MemOS Cloud).
When minimal setup overhead is a critical requirement since self-hosting MemOS necessitates Neo4j and Qdrant for optimal performance.

Pricing:: unknown - Pricing details are not clearly specified in the repository data, but the self-hosted option provides flexibility at the expense of managing infrastructure costs.
Requirements:: Requires Docker; Self-hosting MemOS typically requires setting up Neo4j and Qdrant. The cloud API or OpenClaw Cloud Plugin options do not require these components.

Observed Jul 11, 2026 · Source: enrich:decision_facts

MemOS alternatives Trust report

MemOS vs Memori MemOS vs OpenMemory MemOS vs mem0

Verify the decision

Maintenance and security

Full trust report

Maintenance: Very active (0d since push); As of 1d
Provenance: Not a fork · Organization account; As of 1d
Security (OSV): 2 low (2 low); As of 1mo

Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.

Install

npm install MemOS

npm

How it fits your stack(8)

Typed graph edges - alternatives, integrations, successors, and dependencies. Ranked by relationship type, not raw GitHub stars.

Relationship graph

Optional deeper exploration of typed edges and category neighbours.

AlternativeIntegratesDepends onRelated

AlternativeIntegratesDepends onRelated

Similar tools

Same-category neighbours not already linked as typed edges.

mempalace

The best-benchmarked open-source AI memory system.

58k★

memU

Personal memory for agents with fast retrieval and self-evolving skills

14k★

EverOS

One portable memory layer for every AI agent

12k★

Evidence and technical details

Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.

Overview

MemOS is a self-evolving operating system designed for Long-Short Term Memory (LSTM) systems and AI agents. It supports ultra-persistent memory, hybrid retrieval technologies, and enables the reuse of skills across tasks to save tokens.

Capability facts

Deploy: Self-host
Source: dockerfile:Dockerfile · Aug 18, 2026
Docker: Dockerfile present
Source: dockerfile:Dockerfile · Aug 18, 2026
CLI: CLI entrypoint
Source: pyproject.toml:[project.scripts] · Aug 18, 2026
Languages: typescript, python
Source: github.language+pyproject.toml · Aug 18, 2026

Compatibility

Sourced claims from the README excerpt - not unsourced marketing copy.

Node.js runtimeNode.js

Source: README excerpt (regex_v1, Aug 18, 2026)

| Get an API key | docker compose up | openclaw plugins install | npm install + agent-specific setup |

Source link

README

🚀 Quick Start

MemOS is built around four entry points. Pick the one that matches your scenario.

	Cloud API	Self-Host	MemOS Cloud Plugin	Local Plugin
Best for	Your app, fully managed	Teams on own infra	OpenClaw users, zero ops	DeepSeek Harness, Hermes, or OpenClaw; on-device
Setup	Get an API key	docker compose up	openclaw plugins install	npm install + agent-specific setup
Infra needed	None (hosted)	Neo4j + Qdrant	None (uses MemOS Cloud)	None (local SQLite)
Data lives	MemOS Cloud	Your servers	MemOS Cloud	Your machine

🖥️ Self-Host the MemOS Service

You want to run MemOS as a REST service on your own machine or cluster.

Option A — Docker (recommended):

git clone https://github.com/MemTensor/MemOS.git
cd MemOS
cp docker/.env.example .env          # fill in your API keys in .env
cd docker
docker compose up                    # starts MemOS API + Neo4j + Qdrant

The API is served at http://localhost:8000.

Option B — Run with uvicorn (without Docker):

git clone https://github.com/MemTensor/MemOS.git
cd MemOS
cp docker/.env.example .env          # fill in your API keys in .env

---

## ⚖️ License

MemOS is licensed under the [Apache 2.0 License](./LICENSE).

For agents

This page has a .md twin and JSON over the API.

/tools/memtensor-memos.md GET /api/graphcanon/tools/memtensor-memos GET /api/graphcanon/graph?tool=memtensor-memos