A-MEM: Agentic Memory for LLM Agents

By Wujiang Xu, Zujie Liang, Kai Mei, Hang Gao, Juntao Tan, Yongfeng Zhang. arXiv.org, 2025

752 citations recorded by the source corpus.

While large language model (LLM) agents can effectively use external tools for complex real-world tasks, they require memory systems to leverage historical experiences. Current memory systems enable basic storage and retrieval but lack sophisticated memory organization, despite recent attempts to incorporate graph databases. Moreover, these systems'fixed operations and structures limit their adaptability across diverse tasks. To address this limitation, this paper proposes a novel agentic memory system for LLM agents that can dynamically organize memories in an agentic way. Following the basic principles of the Zettelkasten method, we designed our memory system to create interconnected knowledge networks through dynamic indexing and linking. When a new memory is added, we generate a comprehensive note containing multiple structured attributes, including contextual descriptions, keywords, and tags. The system then analyzes historical memories to identify relevant connections, establishing links where meaningful similarities exist. Additionally, this process enables memory evolution - as new memories are integrated, they can trigger updates to the contextual representations and attributes of existing historical memories, allowing the memory network to continuously refine its understanding. Our approach combines the structured organization principles of Zettelkasten with the flexibility of agent-driven decision making, allowing for more adaptive and context-aware memory management. Empirical experiments on six foundation models show superior improvement against existing SOTA baselines. The source code for evaluating performance is available at https://github.com/WujiangXu/A-mem, while the source code of the agentic memory system is available at https://github.com/WujiangXu/A-mem-sys.

Subfields: Artificial Intelligence, Mechanical Engineering

View Paper: https://www.semanticscholar.org/paper/1f35a15fe9df43d24ec6ea551ec6c9766c17eccf

Citations in the corpus

• AI Agents vs. Agentic AI: A Conceptual Taxonomy, Applications and Challenges (2025, neutral): “This emerging class of systems extends the capabilities of traditional AI Agents by enabling multiple intelligent entities to collaboratively pursue goals through structured communication [146]–[148], shared memory [149], [150], and dynamic role assignment [21].”
• Choosing How to Remember: Adaptive Memory Structures for LLM Agents (2026, unknown): “A-MEM supports hierarchical and relational reasoning by maintaining structured connections between notes, facilitating effective recall during interactions.”
• eMEM: A Hybrid Spatio-Temporal Memory System For Embodied Agents (2026, unknown): “A-MEM [Xu et al., 2025] organises memories as interconnected Zettelkasten notes, and eMEM’s graph achieves similar connectivity but with spatial and temporal grounding.”
• Do Agents Dream of False Memories? Black-box Visual Attacks on Long-term Memory in Multimodal AI Agents (2026, unknown): “These agents rely on multimodal persistent long-term memory: visual and textual content from past interactions are encoded into a shared embedding space, persisted across sessions, and retrieved at query time [86, 54, 7, 71, 43, 73, 57, 45, 18].”
• Constrained Dominant Sets for Multimodal Document Question Answering (2026, unknown): “• The first application of constrained dominant sets to multimodal RAG is demonstrated, instantiated on an agentic-memory-style note graph [26] that incorporates note–note dissimilarity edges, query–note similarity edges, and employs replicator dynamics as the solver.”
• Min-Cost Flow Routing for Evidence Assembly in Long Multimodal Documents (2026, unknown): “G 2 -Reader (Du et al., 2026) maintains two evolving graphs consolidated through VLM message passing, an approach related to graph memories for long-horizon agents (Chhikara et al., 2025; Xu et al., 2025b; Rasmussen et al., 2025).”
• Structured Belief State and the First Precision-Aware Benchmark for LLM Memory Retrieval (2026, unknown): “A-MEM [7] treats memory as a self-organizing network of Zettelkasten-inspired notes where new memories trigger updates to the contextual representations of existing ones.”
• SWE-Adept: An LLM-Based Agentic Framework for Deep Codebase Analysis and Structured Issue Resolution (2026, unknown): “A-MEM (Xu et al., 2025) dynamically integrates prior experiences into a structured graph.”