IAAR-Shanghai/Awesome-AI-Memory
A comprehensive curated repository aggregating research, engineering frameworks, system designs, and benchmarks on AI memory systems for LLMs and agents.

Velocity · 7d
+5.1
★ / day
Trend
→steady
star history
The repository serves as a centralized knowledge base on LLM and agent memory systems. It systematically organizes frontier research on long-term memory, retrieval mechanisms, reasoning over time, and memory-native system design. The collection includes 402 research papers and references 104 open source projects, providing coverage across continual learning, memory-augmented models, and RAG approaches.