Awesome-AI-Memory
Awesome AI Memory | LLM Memory | A curated knowledge base on AI memory for LLMs and agents, covering long-term memory, reasoning, retrieval, and memory-native s
What is Awesome-AI-Memory?
Awesome AI Memory | LLM Memory | A curated knowledge base on AI memory for LLMs and agents, covering long-term memory, reasoning, retrieval, and memory-native system design. Awesome-AI-Memory 是一个 集中式、持续更新的 AI 记忆知识库，系统性整理了与 大模型记忆（LLM Memory）与智能体记忆（Agent Memory） 相关的前沿研究、工程框架、系统设计、评测基准与真实应用实践。
Overview of Awesome-AI-Memory
What it is
Awesome-AI-Memory is a curated knowledge base on AI memory for LLMs and agents. It systematically collects research papers, engineering frameworks, system designs, evaluation benchmarks, and real-world application practices related to long-term memory, reasoning, retrieval, and memory-native system design. The repository aims to connect research directions including natural language processing, information retrieval, agent systems, and cognitive science.
Key features
- A centralized, continuously updated collection of resources on LLM memory and agent memory.
- Categorizes papers into methods & frameworks, datasets & benchmarks, models & systems, and surveys.
- Defines a clear project scope: focuses on agent memory and internal memory mechanisms of language models, excluding pure vision, low-level robot state, hardware/KV cache efficiency, generalized RAG, long context, and tool routing.
- Includes core concept explanations covering memory systems, memory operations, memory management, memory classification, and memory mechanisms.
- Provides a paper list with recent updates (e.g., 222 papers added on 2026-09-15).
Use cases
Researchers and engineers building intelligent systems that require long-term memory, continuous reasoning, and adaptation over time. The repository serves as a reference for understanding and implementing memory architectures for LLMs and agents.
Getting started
See the repository README for instructions on browsing the paper list, contributing, and accessing the resources.
License and pricing
Licensed under Apache-2.0. Open source.
Based on the project's public README and repository metadata.
