Agent_Memory_Techniques
Agent memory for LLMs: 30 runnable Jupyter notebooks covering conversation buffers, vector stores, knowledge graphs, episodic and semantic memory, MemGPT, Mem0,
What is Agent_Memory_Techniques?
Agent memory for LLMs: 30 runnable Jupyter notebooks covering conversation buffers, vector stores, knowledge graphs, episodic and semantic memory, MemGPT, Mem0, Letta, Zep, Graphiti, LoCoMo benchmarks, and production patterns.
Overview of Agent_Memory_Techniques
What it is
Agent_Memory_Techniques is a repository that provides 30 runnable Jupyter notebooks covering agent memory techniques for LLMs. It includes implementations for conversation buffers, vector stores, knowledge graphs, episodic and semantic memory, MemGPT, Mem0, Letta, Zep, Graphiti, LoCoMo benchmarks, and production memory patterns.
Key features
- 30 runnable Jupyter notebooks, each demonstrating a distinct memory technique - Techniques grouped into six families: short-term memory, long-term storage, cognitive architectures, retrieval and routing, frameworks, and evaluation and production - Covers both foundational patterns (e.g., conversation buffer, sliding window) and advanced frameworks (e.g., Mem0, Letta, Zep, Graphiti)
- Includes a taxonomy table and a decision tree to help choose the right technique - Notebooks are designed to be runnable with real code
Use cases
- Managing conversation context within a single chat session - Persisting knowledge across multiple sessions and users - Building cognitive architectures with multiple memory stores - Evaluating and benchmarking memory systems - Deploying memory patterns in production agent systems
Getting started
See the repository README for installation and quick-start instructions.
License and pricing
Licensed under Apache-2.0. Open source.
Based on the project's public README and repository metadata.