mcp-memory-service and nocturne_memory
These are competitors—both provide persistent memory backends for AI agents with graph-based architectures, so you would select one based on preferences around consolidation strategies (A's autonomous consolidation vs. B's rollbackable versioning) rather than using them together.
About mcp-memory-service
doobidoo/mcp-memory-service
Open-source persistent memory for AI agent pipelines (LangGraph, CrewAI, AutoGen) and Claude. REST API + knowledge graph + autonomous consolidation.
This service provides a shared, persistent memory for AI agents, allowing them to retain information and learn across different tasks and sessions. It takes in decisions and facts from various agents, organizing them into a knowledge graph. The output is fast, relevant context that helps agents make better decisions, suitable for anyone building or managing multi-agent AI systems, like AI solution architects or AI product managers.
About nocturne_memory
Dataojitori/nocturne_memory
A lightweight, rollbackable, and visual Long-Term Memory Server for MCP Agents. Say goodbye to Vector RAG and amnesia. Empower your AI with persistent, graph-like structured memory across any model, session, or tool. Drop-in replacement for OpenClaw.
This project offers a long-term memory server that helps AI agents remember who they are and their past experiences across different sessions and models. It takes in structured memory entries, which can be created or updated by the AI itself, and provides a persistent, graph-like knowledge base. This is for developers building and managing AI agents who want their creations to have a continuous, evolving identity rather than starting fresh with each interaction.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work