GH Repository · BAI-LAB
MemoryOS
[EMNLP 2025 Oral] MemoryOS is designed to provide a memory operating system for personalized AI agents.
- stars
- 1,577
- 30-day movement
- +41/day
- Related entries
- 60
- Connections
- 1
dockerretrieval-augmented-generationagent-appragoperating-systemlong-term-memorypersonalizationllmPythonlanguage-modelagent
MemoryOS is a Python library from BAI-LAB that provides an 'operating system' style memory layer for personalized AI agents. It was published as an EMNLP 2025 Oral and focuses on long-term memory with retrieval-augmented generation.
Use it when you want to give LLM agents persistent, personalized long-term memory rather than stateless conversations.
Use it to
- Add long-term memory to a conversational agent
- Personalize agent responses per user
- Integrate retrieval-augmented memory into an LLM pipeline
- Experiment with memory-management architectures for agents
For Developers and researchers building personalized LLM agents
- Role
- agent-app
- Language
- Python
- Licence
- Apache-2.0
- Forks
- 163
- Open issues
- 15
- Last push
- 2026-07-07
- Latest release
- v1.0 · 2025-07-12
topicsagentsllmlong-term-memorypersonalizationragpython