MemOS

An self-evolving memory operating system for LLMs and AI agents, providing persistent memory, hybrid retrieval, and cross-task skill reuse.

MCPDevOpen sourcememoryLLMAI agentspersistent memory

View source →

Type: MCP
Transport: http
Open source: Yes
GitHub Stars: ★ 10.0k
Source: mcp-github
Repository: github.com/MemTensor/MemOS

Overview

MemOS is a memory operating system designed specifically for large language models (LLMs) and AI agents. It unifies storage, retrieval, and management of long-term memory, supporting context-aware and personalized interactions. MemOS offers a unified memory API with support for multimodal memory (text, images, tool traces, etc.), asynchronous memory operations, and feedback-based correction. By using MemOS, AI systems can achieve more efficient, personalized interactions and reduce memory token usage by up to 35.24%. Ideal for developers seeking to enhance AI agent memory capabilities.

Capabilities

▪Unified memory API
▪Multimodal memory support
▪Multiple knowledge base management
▪Asynchronous memory operations
▪Memory feedback and correction

Use cases

Enhancing AI agent memory capabilitiesOptimizing multimodal data processingImproving personalized AI agent interactionsReducing memory token consumption

Setup

Requires: Node环境API Key

npm install @memtensor/memos-local-plugin

This information was compiled by AI from public sources and may contain inaccuracies — please refer to the source.

FAQ

What types of memory does MemOS support?

Supports text, images, tool traces, and other modalities.

How do I get started with MemOS?

Install the plugin and integrate via the API.

Related skills

gemini-cli

Gemini CLI is an open-source AI agent that brings Gemini's powerful capabilities directly into the terminal.

MCP

private-gpt

Provides a complete API layer for local models, supporting RAG, skills, tools, and more.

MCP

susi_alexa_skill

A skill that enables question-and-answer interactions between Alexa and Susi AI.

Agent Skill

scientific-agent-skills

Transform any AI agent into a scientific assistant with 147 ready-to-use research skills.

Agent Skill

planning-with-files

File-based persistent planning skill, ideal for AI coding agents and long-running tasks.

Agent Skill

claude-context

Provides code search capabilities for Claude Code, turning the entire codebase into context.

MCP