MemOS

Overview of MemOS

What it is

MemOS is a memory operating system for LLMs and AI agents. It unifies store, retrieve, and manage operations for long-term memory, enabling context-aware and personalized interactions. It supports knowledge bases, multi-modal data, tool memory, and enterprise-grade optimizations.

Key features

• Unified Memory API: A single API to add, retrieve, edit, and delete memory, structured as a graph.
• Multi-Modal Memory: Supports text, images, tool traces, and personas, retrieved and reasoned together.
• Multi-Cube Knowledge Base Management: Manage multiple knowledge bases as composable memory cubes, enabling isolation and dynamic composition.
• Asynchronous Ingestion via MemScheduler: Run memory operations asynchronously with millisecond-level latency.
• Memory Feedback & Correction: Refine memory with natural-language feedback over time.

Use cases

• AI assistants with consistent, context-rich conversations.
• Customer support that recalls past tickets and user history.
• Personalized agents that adapt to individual preferences.
• Multi-agent collaboration with shared or isolated memory.

Getting started

MemOS provides four entry points: Cloud API, Self-Host, MemOS Cloud Plugin, and Local Plugin. See the repository README for detailed setup instructions. Cloud API Quick Start: python import requests API_KEY = 'mpg-...' base = 'https://memos.memtensor.cn/api/openmem/v1' headers = {'Authorization': f'Token {API_KEY}', 'Content-Type': 'application/json'} # Add a memory requests.post(f'{base}/add/message', headers=headers, json={ 'user_id': 'alice', 'conversation_id': 'conv_001', 'messages': [{'role': 'user', 'content': 'I like strawberry'}] }) # Search memories res = requests.post(f'{base}/search/memory', headers=headers, json={ 'query': 'What do I like?', 'user_id': 'alice' }) print(res.json()) Self-Host (Docker): bash git clone https://github.com/MemTensor/MemOS.git cd MemOS cp docker/.env.example .env cd docker docker compose up

License and pricing

Apache-2.0. Open source. The repository does not state pricing for any commercial or hosted version.

Based on the project's public README and repository metadata.