memobase and MemOS
These are competitors offering different architectural approaches to agent memory—Memobase provides user profile-based long-term memory for chatbots through a dedicated service, while MemOS implements a persistent skill memory OS directly within LLM/agent systems for cross-task reuse.
About memobase
memodb-io/memobase
User Profile-Based Long-Term Memory for AI Chatbot Applications.
Structures user data into dynamically-evolving profiles and timestamped event timelines, enabling sub-100ms memory retrieval through SQL queries rather than vector search. Supports Python, Node.js, and Go SDKs with batch processing buffers to reduce LLM token costs by 40-50%, and includes a Model Context Protocol (MCP) server for seamless integration with AI frameworks. Achieves state-of-the-art performance on the LOCOMO benchmark while maintaining configurable memory schemas, allowing developers to define precisely which user attributes their applications capture.
About MemOS
MemTensor/MemOS
AI memory OS for LLM and Agent systems(moltbot,clawdbot,openclaw), enabling persistent Skill memory for cross-task skill reuse and evolution.
Implements a unified graph-based memory architecture with multi-modal support (text, images, tool traces, personas) and asynchronous ingestion via Redis Streams scheduling, achieving 43.70% accuracy gains over OpenAI Memory while reducing token usage by 35.24%. Integrates natively with OpenClaw agents through both cloud-hosted and local SQLite plugins, featuring hybrid search (FTS5 + vector), automatic task summarization, skill evolution, and natural-language feedback mechanisms for persistent memory refinement.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work