A self-evolving memory operating system for LLM and AI agents
MemTensor/MemOS · skills
Self-evolving memory OS for LLM & AI Agents: ultra-persistent memory, hybrid-retrieval, and cross-task skill reuse, with 35.24% token savings and DeepSeek Harness support.
★ 11,230GitHub stars
1,004Forks
2026-08-24Last updated
TypeScriptLanguage
Apache-2.0License
Key features
- Unified Memory APIA single API to add, retrieve, edit, and delete memory—structured as a graph, inspectable and editable by design, not a black-box embedding store.
- MemOS is a Memory Operating System for LLMs and AI agents that unifies store / retrieve / manage for long-term memory, enabling context-aware and personalized interactions with KB, multi-modal, tool memory, and enterprise-grade optimizations built in.
- 2026-08-17 · MemOS Connects with DeepSeek Harness MemOS now brings persistent memory to DeepSeek Harness through both local and cloud plugins.
- AI assistants with consistent, context-rich conversations
- Multi-Modal MemoryNatively supports text, images, tool traces, and personas, retrieved and reasoned together in one memory system.
Install command
dsh plugin --profile web add github:MemTensor/MemOS