One Memory Layer, Multiple Models (Claude, GPT, Llama, etc.)

Meet MemMachine, an open-source memory layer for advanced AI agents. It enables AI-powered applications to learn, store, and recall data and preferences from past sessions to enrich future interactions. MemMachine's memory layer persists across multiple sessions, agents, and large language models, building a sophisticated, evolving user profile. It transforms AI chatbots into personalized, context-aware AI assistants designed to understand and respond with better precision and depth.

• Developers building AI agents, assistants, or autonomous workflows.
• Researchers experimenting with agent architectures and cognitive models.

• Multiple Memory Types: MemMachine supports Working (Short Term), Persistent (Long Term), and Personalized (Profile) memory types.
• Developer Friendly APIs: Python SDK, RESTful, and MCP interfaces and endpoints to make integrating MemMachine easy into your Agents. For more information, refer to the API Reference Guide.

1. Agents Interact via the API Layer Users interact with an agent, which connects to the MemMachine Memory core through a RESTful API, Python SDK, or MCP Server.
2. MemMachine Manages Memory MemMachine processes interactions and stores them in two distinct types: Episodic Memory for conversational context and Profile Memory for long-term user facts.
3. Data is Persisted to Databases Memory is persisted to a database layer where Episodic Memory is stored in a graph database and Profile Memory is stored in an SQL database.

MemMachine's versatile memory architecture can be applied across any domain, transforming generic bots into specialized, expert assistants. Our growing list of examples showcases the endless possibilities of memory-powered agents that integrate into your own applications and solutions.

• CRM Agent: Your agent can recall a client's entire history and deal stage, proactively helping your sales team build relationships and close deals faster.
• Healthcare Navigator: Offer continuous patient support with an agent that remembers medical history and tracks treatment progress to provide a seamless healthcare journey.
• Personal Finance Advisor: Your agent will remember a user's portfolio and risk tolerance, delivering personalized financial insights based on their complete history.
• Content Writer: Build an assistant that remembers your unique style guide and terminology, ensuring perfect consistency across all documentation.

We're excited to see what you're working on. Join the Discord Server and drop a shout-out to your project in the showcase channel.

Want to get started right away? Check out our Quick Start Guide.

MemMachine is distributed as a Docker container and Python package. For full installation options, visit the documentation.

Get started with a simple "Hello World" example by following the Quick Start Guide.

• Main Website
• Docs & API Reference

• Discord: Join our Docker community for support, updates, and discussions: https://discord.gg/usydANvKqD
• Issues & Feature Requests: Use GitHub Issues.

We welcome contributions! Please see our CONTRIBUTING.md for guidelines.

MemMachine is released under the Apache 2.0 License.