TCCI-backed EverMind Unveils EverMemOS, a Brain-Inspired Long-Term Memory System

Wait 5 sec.

The EverMind team has released EverMemOS, a flagship long-term memory operating system designed for AI agents.Positioned as a world-class data infrastructure for future intelligent systems, EverMemOS aims to give AI agents durable, coherent, and continuously evolving “souls.”In key long-term memory benchmarks, including LoCoMo and LongMemEval-S, the system has already delivered substantial performance gains over prior approaches, establishing itself as the new state of the art.The need for such a solution stems from a structural weakness in today’s large language models: fixed context windows. During long-duration tasks, AI systems frequently “forget,” leading to fragmented memories, factual inconsistencies, and shallow personalization that cannot stand up in real use. Without the ability to retain interaction histories or preserve intermediate reasoning states, AI applications lose coherence—and their practical value diminishes sharply.This limitation has become more than a technical issue; it is now viewed as a fundamental bottleneck in AI’s evolution toward higher intelligence. Without memory, AI systems cannot maintain consistent behavior, agency, or self-iteration over time. Personalization, coherence, and autonomy—all prerequisites for advanced forms of intelligence—depend on a robust, reliable memory architecture.Major AI developers have already embraced this direction. Both Claude and ChatGPT now incorporate long-term memory as a strategic capability, signaling a clear industry trend: memory is emerging as the defining differentiator for the next generation of AI applications. It is also becoming the key mechanism enabling AI’s shift from passive tools to proactive agents capable of long-term adaptation and evolution.While the industry has experimented with early solutions—such as RAG pipelines and emerging memory extensions—these efforts remain largely fragmented. The market still lacks a unified, production-ready memory system that can seamlessly support both deeply personalized experiences and complex multi-user enterprise workflows.What’s missing, EverMind argues, is a platform that integrates accuracy, speed, ease of use, and broad application compatibility at scale. In its absence, developers continue to face an acute unmet need: a high-performance, pluggable, and easily optimized memory layer capable of extending large models into fully empowered AI agents. EverMemOS aims to fill that gap.Inspired by the Human Brain’s Memory MechanismsThe EverMind team originates from the Shanda Group, a tech and investment conglomerate that once led China's wave of digital innovation. Their inspiration comes from the memory mechanisms of the human brain: from encoding sensory signals, hippocampal indexing, to long-term storage in the cortex, with the prefrontal cortex and hippocampus working together to form and retrieve memories. This “brain-like” concept is at the heart of EverMemOS’s design, enabling AI to think, remember, and grow like a human being.This vision also aligns with Shanda founder Chen Tianqiao's longstanding commitment to integrating neuroscience and AI research, demonstrating the vital significance of bringing artificial intelligence and human intelligence together.At the inaugural Symposium for AI Accelerated Science (AIAS 2025), held by the Tianqiao and Chrissy Chen Institute (TCCI) in San Francisco on Oct. 27–28, a leading researcher outlined five core capabilities of what he described as “discovery intelligence,” including long-term memory.He argued that today’s mainstream AI systems are built on a “spatial-structure paradigm” — one that is instantaneous and static, relying on massive parameter scaling to fit snapshots of the world. In contrast, the human brain operates under a “temporal-structure paradigm” that is continuous and dynamic, aiming to manage and predict information as it unfolds over time. Within this framework, long-term memory serves as the essential bridge linking time and intelligence.Inspired by this principle, the team introduced EverMemOS, a system designed to give AI continuity across time. The technology enables models to remember, adapt, and evolve within a temporal stream, representing what the institute views as a foundational step toward more human-like, time-aware artificial intelligence.Against this backdrop, the EverMind team launched EverMemOS, a memory system that has achieved key breakthroughs in both application scenarios and technological performance.In terms of application scenarios: It is the industry's first memory system truly capable of supporting both one-on-one conversations and complex multi-user collaboration. The innovative AI Native product Tanka has already adopted it as an early adopter.In terms of technological performance: Leveraging an innovative bio-inspired “engram” heuristic for memory retrieval and application, EverMemOS achieved impressively high scores on the most prominent long-term memory evaluation datasets: 92.3% on LoCoMo and 82% on LongMemEval-S. Both scores significantly surpass the SOTA (State-of-the-Art) benchmarks, setting new industry standards.EverMemOS Four-Layer ArchitectureInspired by the memory mechanisms of the human brain, EverMemOS introduces an innovative four-layer architecture, drawing analogies to the brain’s key functional areas:Agentic Layer — Responsible for task understanding, decomposition, and generation. This can be compared to the role of the "prefrontal cortex" in attention, planning, and executive control.Memory Layer — Manages the extraction and structured storage of long-term memory, corresponding to the function of the "cerebral cortical network" in long-term consolidation and storage.Index Layer — Connects and efficiently retrieves memories using embeddings, key-value pairs, and knowledge graphs. This resembles the "hippocampus," which is responsible for associating and rapidly indexing memories.API/MCP Interface Layer — Seamlessly integrates with enterprise applications, serving as the AI's sensory interface with the external environment.Three Core Features of EverMemOS:Feature 1: From a “Memory Database” to a “Memory Processor”. EverMemOS’s foremost innovation is that it functions not just as a repository for memory, but as a processor for memory applications. It addresses the core limitation of existing solutions that “just find data without actually using it.” Through its unique reasoning and integration mechanisms, EverMemOS enables memories to actively and instantly shape the model’s thought processes and responses, ensuring that every statement made by the AI is grounded in a long-term understanding of the user. This delivers truly coherent and personalized interactions.Feature 2: Innovative Hierarchical Memory Extraction and Dynamic Organization. At the heart of EverMemOS lies its novel approach to “hierarchical memory extraction.” Instead of treating memory as a chaotic collection of text fragments, it extracts continuous semantic blocks as contextual memory units and dynamically organizes them into structured memories. This layered memory organization links related memories together, overcoming the limitations of pure text-based similarity searches that struggle to capture implicit context, and lays a solid foundation for advanced memory utilization.Feature 3: The Industry’s First Scalable Modular Memory Framework. In real-world applications, memory requirements can vary greatly in different scenarios. To address this, EverMemOS has introduced an innovative, scenario-based scalable memory framework. It flexibly supports various types of memory needs—whether high-precision, structured information is required for work scenarios, or empathy and understanding of implicit emotions are needed for companionship scenarios, EverMemOS can intelligently provide the optimal strategy for organizing and utilizing memory. This effectively overcomes the limitations of traditional single-form memory solutions, meeting the ever-changing needs of diverse environments.Currently, EverMind has released the open-source version of EverMemOS on GitHub for developers and AI teams to deploy and try out. You can access at: https://github.com/EverMind-AI/EverMemOS/. The team plans to launch a cloud service version later this year, offering enterprise users more robust technical support, persistent data storage, and a scalable experience. Interested developers or businesses can leave their email on the official website (http://everm.com) for a chance to be among the first to try the new service.更多精彩内容,关注钛媒体微信号(ID:taimeiti),或者下载钛媒体App