OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
Agent Memory & RAG

Agent Memory Techniques

Agent Memory Techniques is a public collection of 30 Jupyter notebooks that demonstrate memory techniques for LLM agents, including conversation buffers, vector stores, knowledge graphs, episodic and semantic memory, working memory, and production patterns.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Agent Memory Techniques — OpenViST project profile in Agent Memory & RAG.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use it when you want a hands-on learning resource for choosing or prototyping memory patterns in an LLM-agent project, rather than a hosted agent service.

Who it is for

  • Developers building LLM agents who want to compare memory approaches.
  • Learners exploring agent memory through notebook examples.
  • Practitioners looking for examples involving named memory systems and benchmarks.

Documented uses

  • Study and compare conversation-buffer and sliding-window memory.
  • Explore vector-store and knowledge-graph memory patterns.
  • Learn how episodic, semantic and working memory are represented in agent examples.
  • Inspect examples involving MemGPT, Mem0, Letta, Zep, Graphiti and LoCoMo benchmarks.
Before you choose

Requirements & limits

The repository is public and specifies an Apache-2.0 license. Its README says notebooks can be read directly on GitHub or run in the cloud through Colab badges. Local setup calls for Python 3.10 or later, Jupyter, installed requirements and API keys for OpenAI and/or Anthropic examples.

  • The material is delivered as Jupyter notebooks, so local use requires a Python/Jupyter environment or a supported cloud notebook route.
  • The README provides a learning path and decision tree as well as a direct starting notebook.
  • The collection covers both foundational patterns and examples involving named memory systems, benchmarks and production patterns.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up and open the first memory notebook

Prepare the documented local environment and launch the conversation-buffer notebook.

Before you start

  • Python 3.10 or later
  • Git and a command-line shell
  • Jupyter
  • OpenAI and/or Anthropic API key for the configured examples

Steps

  1. Clone the repository and change into its directory.

  2. Create and activate a Python virtual environment.

  3. Install dependencies from requirements.txt.

  4. Copy .env.example to .env and enter the relevant API key or keys.

  5. Launch Jupyter for all_techniques/01_conversation_buffer_memory/.

Expected outcome

A local Jupyter session opened at the conversation-buffer memory notebook.

Guide limitations

  • The instructions do not establish that every notebook uses the same provider or credentials.
  • The API keys are prerequisites for the configured examples; the repository can also be read in a browser or opened using its Colab badges.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
9 October 2026
Original website title
NirDiamant/Agent_Memory_Techniques · Official homepage title · retained in its source language
How we review sourcesSuggest a correction