<p>/PRNewswire/ -- 人工智能基础设施公司 EverMind 宣布在长期记忆研究方面取得了重大里程碑，其新发布的 EverMemOS 在 LoCoMo 上达到了 92.3% 的准确率，在 LongMemEval-S 上达到了 82% 的准确率——这两个都是该领域最权威的基准测试。结果显著超越了之前的方法，并为持久、连贯的机器记忆建立了新的最先进标准。EverMemOS 的开发使命是为未来的智能构建记忆层，目前已为 Tanka 提供支持，并旨在为更多的人工智能代理提供稳定的身份，使其随着时间的推移而演变，从而使它们能够发展出持久、连贯、不断增长的 “灵魂”。</p>
<p>继续阅读</p>
<p><img src="https://imageproxy.pbkrs.com/https://www.prnewswire.com/news-releases/null?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://www.prnewswire.com/news-releases/null"/></p>
<p>EverMind 网站</p>
<p><strong>受人类认知启发的范式转变</strong></p>
<p>EverMemOS 的核心设计源于企业家和慈善家陈天桥在今年十月于旧金山举行的天桥与陈克里斯研究所人工智能加速科学研讨会（AIAS 2025）上分享的见解。陈天桥将当前人工智能的 “空间结构” 范式（瞬时、静态、依赖于大量空间参数）与人脑的 “时间结构” 范式（连续、动态、旨在管理基于时间的信息）进行了对比。他认为，长期记忆是连接时间和真正智能的关键。EverMemOS 正是基于这一理念而诞生：一个旨在赋予人工智能时间连续性的系统——记住、适应和随着时间推移而成长的能力。</p>
<p><strong>重新定义人工智能记忆的三项创新</strong></p>
<p>EverMemOS 通过三项行业首创的创新脱颖而出，解决了长期以来的记忆局限性。</p>
<p>其首项创新是将记忆从静态存储转变为一种主动的、应用的能力。传统方法检索信息但不将其整合到推理中。相比之下，EverMemOS 通过独特的融合和决策机制，使记忆能够持续影响模型的思维和输出。每次交互都基于上下文历史，从而在时间上提供一致、稳定和个性化的体验。</p>
<p>第二项创新是其层次化记忆提取和动态组织引擎。EverMemOS 不再将记忆存储为无结构的文本片段，而是将其转换为语义 MemCells，并在不断发展的记忆图中进行组织。这一策略将相关思想联系起来，克服了基于相似性检索的局限性，并为复杂的下游应用（如推理驱动的代理、专业助手和情感意识的伴侣模型）提供了结构化基础。</p>
<p>第三项创新在于 EverMemOS 的可扩展模块化记忆框架，这是首个旨在满足多样化现实需求的框架。不同场景下的记忆需求各不相同：专业任务需要精确、结构化的信息；伴侣模型需要富有同情心、情感意识的上下文；而代理工作流程则依赖于连续性和任务特定的回忆。EverMemOS 针对每种情况进行适应，选择最佳的记忆组织和应用策略。这种灵活性有助于解决由僵化的单一形式记忆系统带来的长期挑战，使人工智能能够在多样化的环境中有效运作。</p>
<p><strong>为每位用户提供灵活的服务模型</strong></p>
<p>EverMind 将以多种服务格式提供 EverMemOS，以适应不同的技术能力水平。目前，开源版本允许经验丰富的开发者独立部署和维护该系统。今年晚些时候，将发布云服务版本，提供免费试用和基于订阅的访问，无需运营开销。定制部署版本将支持寻求量身定制集成的战略合作伙伴。个人版本则为硬件集成和 MCP 启用的环境而构建，将长期记忆能力扩展到个人用例。</p>
<p>EverMind 正在通过解决人工智能最根本的局限性之一——缺乏长期记忆，来推动人工智能的基础发展。通过 EverMemOS，该公司引入了一种可扩展、可定制的记忆架构，使人工智能能够在扩展上下文中运作，保持行为一致性，并通过持续互动不断改进。随着人工智能系统朝着更大的持久性、适应性和身份形成发展，EverMemOS 被设计为支持下一代智能代理的可靠记忆层。</p>
<p><strong>关于 EverMind</strong></p>
<p>EverMind 正在通过解决人工智能最根本的局限性之一——长期记忆，重新定义人工智能的未来。其旗舰平台 EverMemOS 引入了一种突破性的架构，用于可扩展和可定制的记忆系统，使人工智能能够在扩展上下文中运作，保持行为一致性，并通过持续互动不断改进。</p>
<p>要了解更多关于 EverMind 和 EverMemOS 的信息，请访问：</p>
<p>网站：https://evermind.ai/</p>
<p>GitHub：https://github.com/EverMind-AI/EverMemOS</p>
<p>X：https://x.com/EverMindAI</p>
<p>Reddit：https://www.reddit.com/r/EverMindAI/</p>
<p>来源 EverMind</p>
<p><img src="https://imageproxy.pbkrs.com/https://rt.prnewswire.com/rt.gif/query-RGF0ZUlkPTIwMjUxMjA5Jk5ld3NJdGVtSWQ9Q040MjI4MCZUcmFuc21pc3Npb25fSWQ9MjAyNTEyMDkwMzExUFJfTkVXU19VU1BSX19fX19DTjQyMjgw?x-oss-process=image/auto-orient,1/interlace,1/resize,w_1440,h_1440/quality,q_95/format,jpg" alt="" original-src="https://imageproxy.pbkrs.com/https://rt.prnewswire.com/rt.gif/query-RGF0ZUlkPTIwMjUxMjA5Jk5ld3NJdGVtSWQ9Q040MjI4MCZUcmFuc21pc3Npb25fSWQ9MjAyNTEyMDkwMzExUFJfTkVXU19VU1BSX19fX19DTjQyMjgw"/></p>

<p>EverMind 宣布推出 EverMemOS，这是一项突破性的 AI 记忆技术，在 LoCoMo 基准测试中实现了 92.3% 的准确率，在 LongMemEval-S 基准测试中实现了 82% 的准确率。EverMemOS 引入了 AI 记忆的创新，使 AI 代理能够拥有持久、一致和不断发展的 “灵魂”。它将记忆从静态存储转变为主动能力，采用层次化记忆提取，并提供模块化框架。EverMind 计划为 EverMemOS 提供多种服务形式，以解决 AI 的长期记忆限制，并支持下一代智能代理</p>

<p>, /PRNewswire/ -- AI infrastructure company EverMind has announced a major milestone in long-term memory research with its newly released EverMemOS achieving 92.3% accuracy on LoCoMo and 82% on LongMemEval-S—two of the field's most authoritative benchmarks. The results significantly surpass prior methods and establish a new State-of-the-Art for durable, coherent machine memory. Developed with the mission of building the memory layer for future intelligence, EverMemOS now powers Tanka and is designed to equip more AI agents with stable identities that evolve through time, enabling them to develop durable, coherent, continuously growing &#34;souls.&#34;</p>
<div class="lb-trans"><p>/PRNewswire/ -- 人工智能基础设施公司 EverMind 宣布在长期记忆研究方面取得了重大里程碑，其新发布的 EverMemOS 在 LoCoMo 上达到了 92.3% 的准确率，在 LongMemEval-S 上达到了 82% 的准确率——这两个都是该领域最权威的基准测试。结果显著超越了之前的方法，并为持久、连贯的机器记忆建立了新的最先进标准。EverMemOS 的开发使命是为未来的智能构建记忆层，目前已为 Tanka 提供支持，并旨在为更多的人工智能代理提供稳定的身份，使其随着时间的推移而演变，从而使它们能够发展出持久、连贯、不断增长的 “灵魂”。</p>
</div><p>Continue Reading</p>
<div class="lb-trans"><p>继续阅读</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://www.prnewswire.com/news-releases/null" alt="" original-src="https://www.prnewswire.com/news-releases/null"/></p>
<p>EverMind Website</p>
<div class="lb-trans"><p>EverMind 网站</p>
</div><p><strong>A Paradigm Shift Inspired by Human Cognition</strong></p>
<div class="lb-trans"><p><strong>受人类认知启发的范式转变</strong></p>
</div><p>EverMemOS's core design stems from an insight shared by entrepreneur and philanthropist Chen Tianqiao at the Tianqiao and Chrissy Chen Institute's Symposium for AI Accelerated Science (AIAS 2025) in San Francisco this October. Chen contrasted current AI's &#34;spatial structure&#34; paradigm, which is instantaneous, static, and reliant on massive spatial parameters, with the human brain's &#34;temporal structure&#34; paradigm: continuous, dynamic, and built to manage time-based information. Long-term memory, he argued, is the linchpin connecting time and true intelligence. EverMemOS was born from this idea: a system engineered to give AI temporal continuity—the ability to remember, adapt, and grow over time.</p>
<div class="lb-trans"><p>EverMemOS 的核心设计源于企业家和慈善家陈天桥在今年十月于旧金山举行的天桥与陈克里斯研究所人工智能加速科学研讨会（AIAS 2025）上分享的见解。陈天桥将当前人工智能的 “空间结构” 范式（瞬时、静态、依赖于大量空间参数）与人脑的 “时间结构” 范式（连续、动态、旨在管理基于时间的信息）进行了对比。他认为，长期记忆是连接时间和真正智能的关键。EverMemOS 正是基于这一理念而诞生：一个旨在赋予人工智能时间连续性的系统——记住、适应和随着时间推移而成长的能力。</p>
</div><p><strong>Three Innovations Redefining AI Memory</strong></p>
<div class="lb-trans"><p><strong>重新定义人工智能记忆的三项创新</strong></p>
</div><p>EverMemOS stands out through three industry-first innovations that solve longstanding memory limitations.</p>
<div class="lb-trans"><p>EverMemOS 通过三项行业首创的创新脱颖而出，解决了长期以来的记忆局限性。</p>
</div><p>Its first innovation is a shift from treating memory as static storage to enabling it as an active, applied capability. Conventional methods retrieve information but do not integrate it into reasoning. EverMemOS, by contrast, uses unique fusion and decision mechanisms that allow memory to continuously influence a model's thinking and outputs. Each interaction draws on contextual history, enabling consistent, stable, and personalized experiences across time.</p>
<div class="lb-trans"><p>其首项创新是将记忆从静态存储转变为一种主动的、应用的能力。传统方法检索信息但不将其整合到推理中。相比之下，EverMemOS 通过独特的融合和决策机制，使记忆能够持续影响模型的思维和输出。每次交互都基于上下文历史，从而在时间上提供一致、稳定和个性化的体验。</p>
</div><p>The second innovation is its hierarchical memory extraction and dynamic organization engine. Rather than storing memories as unstructured text fragments, EverMemOS converts them into semantic MemCells and organizes them within evolving memory graphs. This strategy links related ideas, overcomes the limitations of similarity-based retrieval, and provides a structured foundation for complex downstream applications such as reasoning-driven agents, professional assistants, and emotionally aware companion models.</p>
<div class="lb-trans"><p>第二项创新是其层次化记忆提取和动态组织引擎。EverMemOS 不再将记忆存储为无结构的文本片段，而是将其转换为语义 MemCells，并在不断发展的记忆图中进行组织。这一策略将相关思想联系起来，克服了基于相似性检索的局限性，并为复杂的下游应用（如推理驱动的代理、专业助手和情感意识的伴侣模型）提供了结构化基础。</p>
</div><p>A third innovation lies in EverMemOS's extensible modular memory framework, the first of its kind designed to meet diverse real-world needs. Memory requirements differ across scenarios: professional tasks demand precise, structured information; companionship models require empathetic, emotionally aware context; and agentic workflows rely on continuity and task-specific recall. EverMemOS adapts to each case, selecting the optimal memory organization and application strategy. This flexibility helps address the long-standing challenge posed by rigid, single-form memory systems and enables AI to operate effectively across varied environments.</p>
<div class="lb-trans"><p>第三项创新在于 EverMemOS 的可扩展模块化记忆框架，这是首个旨在满足多样化现实需求的框架。不同场景下的记忆需求各不相同：专业任务需要精确、结构化的信息；伴侣模型需要富有同情心、情感意识的上下文；而代理工作流程则依赖于连续性和任务特定的回忆。EverMemOS 针对每种情况进行适应，选择最佳的记忆组织和应用策略。这种灵活性有助于解决由僵化的单一形式记忆系统带来的长期挑战，使人工智能能够在多样化的环境中有效运作。</p>
</div><p><strong>Flexible Service Models for Every User</strong></p>
<div class="lb-trans"><p><strong>为每位用户提供灵活的服务模型</strong></p>
</div><p>EverMind will offer EverMemOS in multiple service formats to fit different levels of technical capability. At present, an open-source version allows experienced developers to deploy and maintain the system independently. Later this year, a cloud-service version will be released to offer free trials and subscription-based access, with no operational overhead. A custom deployment version will support strategic partners seeking tailored integrations. A personal version, built for hardware-integrated and MCP-enabled contexts, will extend long-term memory capabilities to individual use cases.</p>
<div class="lb-trans"><p>EverMind 将以多种服务格式提供 EverMemOS，以适应不同的技术能力水平。目前，开源版本允许经验丰富的开发者独立部署和维护该系统。今年晚些时候，将发布云服务版本，提供免费试用和基于订阅的访问，无需运营开销。定制部署版本将支持寻求量身定制集成的战略合作伙伴。个人版本则为硬件集成和 MCP 启用的环境而构建，将长期记忆能力扩展到个人用例。</p>
</div><p>EverMind is working to advance the foundation of artificial intelligence by addressing one of its most fundamental limitations: the lack of long-term memory. Through EverMemOS, the company introduces a scalable, customizable memory architecture that enables AI to operate with extended context, maintain behavioral consistency, and improve through continuous interaction. As AI systems move toward greater persistence, adaptability, and identity formation, EverMemOS is designed to serve as a reliable memory layer supporting the next generation of intelligent agents.</p>
<div class="lb-trans"><p>EverMind 正在通过解决人工智能最根本的局限性之一——缺乏长期记忆，来推动人工智能的基础发展。通过 EverMemOS，该公司引入了一种可扩展、可定制的记忆架构，使人工智能能够在扩展上下文中运作，保持行为一致性，并通过持续互动不断改进。随着人工智能系统朝着更大的持久性、适应性和身份形成发展，EverMemOS 被设计为支持下一代智能代理的可靠记忆层。</p>
</div><p><strong>About EverMind</strong></p>
<div class="lb-trans"><p><strong>关于 EverMind</strong></p>
</div><p>EverMind is redefining the future of AI by solving one of its most fundamental limitations: long-term memory. Its flagship platform, EverMemOS, introduces a breakthrough architecture for scalable and customizable memory systems, enabling AI to operate with extended context, maintain behavioral consistency, and improve through continuous interaction.</p>
<div class="lb-trans"><p>EverMind 正在通过解决人工智能最根本的局限性之一——长期记忆，重新定义人工智能的未来。其旗舰平台 EverMemOS 引入了一种突破性的架构，用于可扩展和可定制的记忆系统，使人工智能能够在扩展上下文中运作，保持行为一致性，并通过持续互动不断改进。</p>
</div><p>To learn more about EverMind and EverMemOS, please visit:</p>
<div class="lb-trans"><p>要了解更多关于 EverMind 和 EverMemOS 的信息，请访问：</p>
</div><p>Website: https://evermind.ai/</p>
<div class="lb-trans"><p>网站：https://evermind.ai/</p>
</div><p>GitHub: https://github.com/EverMind-AI/EverMemOS</p>
<div class="lb-trans"><p>GitHub：https://github.com/EverMind-AI/EverMemOS</p>
</div><p>X: https://x.com/EverMindAI</p>
<div class="lb-trans"><p>X：https://x.com/EverMindAI</p>
</div><p>Reddit: https://www.reddit.com/r/EverMindAI/</p>
<div class="lb-trans"><p>Reddit：https://www.reddit.com/r/EverMindAI/</p>
</div><p>SOURCE EverMind</p>
<div class="lb-trans"><p>来源 EverMind</p>
</div><p><img src="https://imageproxy.pbkrs.com/https://rt.prnewswire.com/rt.gif/query-RGF0ZUlkPTIwMjUxMjA5Jk5ld3NJdGVtSWQ9Q040MjI4MCZUcmFuc21pc3Npb25fSWQ9MjAyNTEyMDkwMzExUFJfTkVXU19VU1BSX19fX19DTjQyMjgw" alt="" original-src="https://rt.prnewswire.com/rt.gif?NewsItemId=CN42280&amp;Transmission_Id=202512090311PR_NEWS_USPR_____CN42280&amp;DateId=20251209"/></p>

AI 基础设施公司 EverMind 的 EverMemOS 旨在为 AI 代理提供持久、一致且不断进化的 “灵魂”