{"@context":"https://schema.org","@type":"NewsArticle","generatedAt":"2026-07-23T06:40:50.084Z","headline":"Google Cloud 推出 Always-On Memory Agent，用 LLM 持续整合替代 RAG 与向量数据库","description":"Google Cloud 发布 Always-On Memory Agent，一个基于 ADK 和 Gemini 3.1 Flash-Lite 的 24/7 后台进程，不使用向量数据库或嵌入向量，而是由 LLM 将结构化记忆写入 SQLite。该方案每 30 分钟自动整合记忆并发现关联，支持 27 种文件类型，通过 HTTP API 或 Streamlit 仪表盘交互。","url":"https://www.aioga.com/news/cmrq3c6fz0083bi6vbya5443p/","mainEntityOfPage":"https://www.aioga.com/news/cmrq3c6fz0083bi6vbya5443p/","datePublished":"2026-07-18T07:57:51.000Z","dateModified":"2026-07-18T07:57:51.000Z","inLanguage":"zh-CN","publisher":{"@type":"NewsMediaOrganization","name":"Aioga","url":"https://www.aioga.com"},"citation":["https://www.marktechpost.com/2026/07/18/google-clouds-always-on-memory-agent-replaces-rag-and-embeddings-with-continuous-llm-consolidation-on-gemini-3-1-flash-lite","https://aihot.virxact.com/items/cmrq3c6fz0083bi6vbya5443p"],"canonicalUrl":"https://www.aioga.com/news/cmrq3c6fz0083bi6vbya5443p/","directAnswer":{"@type":"Answer","text":"Aioga 编辑摘要：Google Cloud 发布 Always-On Memory Agent，一个基于 ADK 和 Gemini 3.1 Flash-Lite 的 24/7 Aioga 将其归入「产品更新」方向，重点关注它对真实使用和行业竞争的影响。","url":"https://www.aioga.com/news/cmrq3c6fz0083bi6vbya5443p/","dateCreated":"2026-07-18T07:57:51.000Z","author":{"@type":"Organization","@id":"https://www.aioga.com/authors/aioga-editorial/#editorial-team","name":"Aioga Editorial Team","url":"https://www.aioga.com/authors/aioga-editorial/"}},"evidence":[{"@type":"CreativeWork","name":"marktechpost.com source article","url":"https://www.marktechpost.com/2026/07/18/google-clouds-always-on-memory-agent-replaces-rag-and-embeddings-with-continuous-llm-consolidation-on-gemini-3-1-flash-lite","datePublished":"2026-07-18T07:57:51.000Z","provider":{"@type":"Organization","name":"marktechpost.com","url":"https://www.marktechpost.com/2026/07/18/google-clouds-always-on-memory-agent-replaces-rag-and-embeddings-with-continuous-llm-consolidation-on-gemini-3-1-flash-lite"}},{"@type":"CreativeWork","name":"AIHot archive record","url":"https://aihot.virxact.com/items/cmrq3c6fz0083bi6vbya5443p","datePublished":"2026-07-18T07:57:51.000Z","provider":{"@type":"Organization","name":"AIHot","url":"https://aihot.virxact.com/items/cmrq3c6fz0083bi6vbya5443p"}}],"aggregationSource":"MarkTechPost（RSS）","originalPublisher":{"name":"marktechpost.com","url":"https://www.marktechpost.com/2026/07/18/google-clouds-always-on-memory-agent-replaces-rag-and-embeddings-with-continuous-llm-consolidation-on-gemini-3-1-flash-lite"},"article":{"id":"cmrq3c6fz0083bi6vbya5443p","slug":"cmrq3c6fz0083bi6vbya5443p","url":"https://www.aioga.com/news/cmrq3c6fz0083bi6vbya5443p/","title":"Google Cloud 推出 Always-On Memory Agent，用 LLM 持续整合替代 RAG 与向量数据库","title_en":"Google Cloud's Always-On Memory Agent Replaces RAG and Embeddings With Continuous LLM Consolidation on Gemini 3.1 Flash-Lite","summary":"Google Cloud 发布 Always-On Memory Agent，一个基于 ADK 和 Gemini 3.1 Flash-Lite 的 24/7 后台进程，不使用向量数据库或嵌入向量，而是由 LLM 将结构化记忆写入 SQLite。该方案每 30 分钟自动整合记忆并发现关联，支持 27 种文件类型，通过 HTTP API 或 Streamlit 仪表盘交互。","source":"MarkTechPost（RSS）","sourceUrl":"https://www.marktechpost.com/2026/07/18/google-clouds-always-on-memory-agent-replaces-rag-and-embeddings-with-continuous-llm-consolidation-on-gemini-3-1-flash-lite","aiHotUrl":"https://aihot.virxact.com/items/cmrq3c6fz0083bi6vbya5443p","publishedAt":"2026-07-18T07:57:51.000Z","category":"产品更新","score":62,"selected":false,"articleBody":["Most AI agents forget. They process a request, answer it, then drop the context. Google Cloud’s generative-ai repository now ships a sample that tackles this directly. It is the Always-On Memory Agent ：https://github.com/GoogleCloudPlatform/generative-ai/tree/main/gemini/agents/always-on-memory-agent, a reference implementation that treats memory as a running process.","Fundamentally, the project is a lightweight background agent that never stops. It runs 24/7 as a continuous process, not a one-shot call. It is built with Google ADK (Agent Development Kit) and Gemini 3.1 Flash-Lite . Notably, it uses no vector database and no embeddings. Instead, an LLM reads, thinks, and writes structured memory into SQLite . The model choice targets low latency and low cost for continuous background work.","Architecturally, an orchestrator routes every request to one of three specialist sub-agents. Each sub-agent owns its own tools for reading or writing the memory store.","First, the IngestAgent handles incoming content. It uses Gemini’s multimodal capabilities to extract a summary, entities, topics, and an importance score. That structured record then lands in the memories table.","Next, the ConsolidateAgent runs on a timer, every 30 minutes by default. Like sleep cycles, it reviews unconsolidated memories and finds connections between them. Then it writes a synthesized summary, one key insight, and those connections to the database. Consequently, the agent builds new understanding while idle, with no prompt.","Finally, the QueryAgent answers questions. It reads all memories and consolidation insights, then synthesizes a response. Importantly, it cites the memory IDs it used as sources.","Beyond text, the IngestAgent accepts 27 file types across five categories. Simply drop any supported file into the ./inbox folder for automatic pickup.","To clarify the difference, it frames three common memory approaches. Each solves part of the problem, yet leaves a gap.","Unlike RAG, this agent processes memory actively, not only on retrieval.","Practically, the pattern fits any workload needing durable, evolving context. Consider three examples.","With the design clear, setup stays minimal for early-level engineers. Install dependencies, set your key, then start the process.","Once running, the agent watches ./inbox , consolidates every 30 minutes, and serves an HTTP API on port 8888. Therefore, you can also feed it over HTTP.","Additionally, the API exposes /status , /memories , /consolidate , /delete , and /clear . An optional Streamlit dashboard adds ingest, query, browse, and delete controls. CLI flags change the watch folder, port, and consolidation interval.","Need to partner with us for promoting your GitHub Repo OR Hugging Face Page OR Product Release OR Webinar etc.? Connect with us ：https://forms.gle/wbash1wF6efRj8G58","Michal Sutter is a data science professional with a Master of Science in Data Science from the University of Padova. With a solid foundation in statistical analysis, machine learning, and data engineering, Michal excels at transforming complex datasets into actionable insights.","Build an Agentic Event Venue Operator [Full Codes]：https://pxllnk.co/twdn5","Thanks! Our team will contact you soon"],"articleImages":[{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2025/07/a-professional-linkedin-headshot-photogr_0jcmb0R9Sv6nW5XK-zkPHw_uARV5VW1ST6osLNlunoVWg-300x300.png","alt":"","afterParagraph":13,"url":"/media/articles/cmrq3c6fz0083bi6vbya5443p/84e64b03066de40c.webp"},{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2026/07/blog19132-36-100x70.png","alt":"Perplexity AI Releases WANDR","afterParagraph":14,"url":"/media/articles/cmrq3c6fz0083bi6vbya5443p/c5dd5ffe944a5f8b.webp"},{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2026/07/blog19132-35-100x70.png","alt":"10 Open-Source No-Code Platforms for Building LLM Apps, RAG Systems, and AI Agents","afterParagraph":14,"url":"/media/articles/cmrq3c6fz0083bi6vbya5443p/330c02f0ca218ba6.webp"},{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2026/07/blog19132-34-100x70.png","alt":"Kimi K3 vs DeepSeek V4 Pro vs GLM-5.2","afterParagraph":14,"url":"/media/articles/cmrq3c6fz0083bi6vbya5443p/a80f941097ee6a50.webp"},{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2026/07/blog19132-33-100x70.png","alt":"Fine-Tuning Qwen3 with LoRA Using NVIDIA NeMo AutoModel","afterParagraph":14,"url":"/media/articles/cmrq3c6fz0083bi6vbya5443p/6fe6f7b6366a31dc.webp"},{"sourceUrl":"https://www.marktechpost.com/wp-content/uploads/2026/07/blog19132-32-100x70.png","alt":"NVIDIA Released DeepStream 9.1","afterParagraph":14,"url":"/media/articles/cmrq3c6fz0083bi6vbya5443p/1e950c77753bb01b.webp"}],"mediaStatus":"ok","articleBodyZh":["大多数 AI 代理会遗忘。它们处理请求后给出答案，然后丢掉上下文。Google Cloud 的生成式 AI 仓库现在提供了一个直接解决这一问题的示例。它就是 Always-On Memory Agent：https://github.com/GoogleCloudPlatform/generative-ai/tree/main/gemini/agents/always-on-memory-agent，一个将记忆视为持续运行进程的参考实现。","从根本上说，该项目是一个轻量级的后台代理，永远不会停止。它作为一个连续运行的进程 24/7 工作，而不是一次性调用。它是使用 Google ADK（代理开发工具包）和 Gemini 3.1 Flash-Lite 构建的。值得注意的是，它不使用向量数据库也不使用嵌入。取而代之的是，LLM 会读取、思考，并将结构化记忆写入 SQLite。模型的选择目标是实现低延迟和低成本的连续后台工作。","在架构上，一个编排器将每个请求分配给三个专业子代理之一。每个子代理拥有自己的工具，用于读取或写入记忆存储。","首先，IngestAgent 处理传入内容。它利用 Gemini 的多模态能力提取摘要、实体、主题和重要性评分。然后，该结构化记录被存入 memories 表中。","接下来，ConsolidateAgent 按计时器运行，默认每 30 分钟执行一次。像睡眠周期一样，它会审查未整合的记忆并寻找它们之间的联系。然后，它将一个综合摘要、一项关键见解和这些联系写入数据库。因此，该代理在空闲时也能构建新理解，无需任何提示。","最后，QueryAgent 回答问题。它读取所有记忆和整合见解，然后综合生成答案。重要的是，它会引用所用记忆的 ID 作为来源。","除了文本，IngestAgent 可接受五类共 27 种文件类型。只需将任何受支持的文件放入 ./inbox 文件夹，即可自动采集。","为了澄清区别，它描述了三种常见的记忆方法。每种方法都解决了部分问题，但仍存在空白。","与 RAG 不同，该代理主动处理记忆，而不仅仅是在检索时。","实际上，这种模式适用于任何需要持久、可演化上下文的工作负载。考虑三个示例。","设计完成后，对于初级工程师来说，设置保持最小化。安装依赖项，设置你的密钥，然后启动进程。","运行后，代理会监控 ./inbox，每 30 分钟进行一次整合，并在端口 8888 提供 HTTP API。因此，你也可以通过 HTTP 向它提供数据。","此外，API 还暴露了 /status、/memories、/consolidate、/delete 和 /clear。可选的 Streamlit 仪表板提供数据输入、查询、浏览和删除控制。CLI 标志可以更改监控文件夹、端口和整合间隔。","需要与我们合作推广你的 GitHub 仓库或 Hugging Face 页面或产品发布或网络研讨会等？请联系：https://forms.gle/wbash1wF6efRj8G58","Michal Sutter 是一名数据科学专业人士，拥有帕多瓦大学数据科学硕士学位。凭借在统计分析、机器学习和数据工程方面的坚实基础，Michal 擅长将复杂数据集转化为可执行的见解。","构建一个代理式活动场地运营商 [完整代码]：https://pxllnk.co/twdn5","谢谢！我们的团队会很快联系你"],"translationStatus":"translated","bodyOrigin":"source-page","editorial":{"summary":"Aioga 编辑摘要：Google Cloud 发布 Always-On Memory Agent，一个基于 ADK 和 Gemini 3.1 Flash-Lite 的 24/7 Aioga 将其归入「产品更新」方向，重点关注它对真实使用和行业竞争的影响。","background":"背景分析：产品与工具类动态的价值取决于它是否解决明确场景、能否进入工作流，以及交付、价格和数据安全是否可接受。","viewpoint":"Aioga 判断：这条动态更适合作为行业观察信号，当前信息足以建立线索，但不足以推导长期结论。","implications":"影响分析：对相关团队而言，短期应先核对来源、可用范围和实际成本，再判断是否值得接入或跟进。","nextStep":"后续观察：继续观察产品是否开放使用、用户反馈、定价、集成能力和后续版本更新。","evidenceRefs":["title","summary","articleBody"],"confidence":"medium","status":"published","aiGenerated":false,"autoApproved":true,"generatedBy":"rule-safe-fallback","generatedAt":"2026-07-23T06:49:19.155Z","sourceHash":"b9a9e635e20c81ae","validation":{"passed":true,"mode":"rule-safe-fallback","checks":["schema","length","source-attribution","no-html"]}},"tags":["产品更新","MarkTechPost（RSS）"],"translations":{"zh-CN":{"title":"Google Cloud 推出 Always-On Memory Agent，用 LLM 持续整合替代 RAG 与向量数据库","summary":"Google Cloud 发布 Always-On Memory Agent，一个基于 ADK 和 Gemini 3.1 Flash-Lite 的 24/7 后台进程，不使用向量数据库或嵌入向量，而是由 LLM 将结构化记忆写入 SQLite。该方案每 30 分钟自动整合记忆并发现关联，支持 27 种文件类型，通过 HTTP API 或 Streamlit 仪表盘交互。","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud 推出 Always-On Memory Agent，用 LLM 持续整合替代 RAG 与向量数据库 - Aioga AI资讯","description":"Google Cloud 发布 Always-On Memory Agent，一个基于 ADK 和 Gemini 3.1 Flash-Lite 的 24/7 后台进程，不使用向量数据库或嵌入向量，而是由 LLM 将结构化记忆写入 SQLite。该方案每 30 分钟自动整合记忆并发现关联，支持 27 种文件类型，通过 HTTP API 或 Streamlit ","url":"https://www.aioga.com/news/cmrq3c6fz0083bi6vbya5443p/"},"en":{"title":"Google Cloud launches Always-On Memory Agent, using LLM for continuous integration as an alternative to RAG and vector databases","summary":"Google Cloud has released the Always-On Memory Agent, a 24/7 background process based on ADK and Gemini 3.1 Flash-Lite. It does not use a vector database or embedding vectors, but instead uses an LLM to write structured memory into SQLite. This solution automatically consolidates memory and discovers associations every 30 minutes, supports 27 file types, and can be interacted with via HTTP API or a Streamlit dashboard.","category":"Products","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud launches Always-On Memory Agent, using LLM for continuous integration as an alternative to RAG and vector databases - Aioga AI News","description":"Google Cloud has released the Always-On Memory Agent, a 24/7 background process based on ADK and Gemini 3.1 Flash-Lite. It does not use a vector database or embedding vectors, but...","url":"https://www.aioga.com/en/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:02:31.481Z"},"ja":{"title":"Google Cloud、Always-On Memory Agent を導入、LLM で RAG とベクトルデータベースの代替を継続的に統合","summary":"Google Cloud は Always-On Memory Agent を発表しました。これは ADK と Gemini 3.1 Flash-Lite に基づく 24 時間稼働のバックグラウンドプロセスで、ベクトルデータベースや埋め込みベクトルを使用せず、LLM により構造化メモリを SQLite に書き込みます。このソリューションは 30 分ごとにメモリを自動で統合して関連性を発見し、27 種類のファイルタイプをサポートし、HTTP API または Streamlit ダッシュボードを介して操作できます。","category":"製品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud、Always-On Memory Agent を導入、LLM で RAG とベクトルデータベースの代替を継続的に統合 - Aioga AIニュース","description":"Google Cloud は Always-On Memory Agent を発表しました。これは ADK と Gemini 3.1 Flash-Lite に基づく 24 時間稼働のバックグラウンドプロセスで、ベクトルデータベースや埋め込みベクトルを使用せず、LLM により構造化メモリを SQLite に書き込みます。このソリューションは 30 分ごとにメモ...","url":"https://www.aioga.com/ja/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:02:36.781Z"},"ko":{"title":"Google Cloud는 Always-On Memory Agent를 출시하여 LLM으로 RAG 및 벡터 데이터베이스를 지속적으로 통합합니다","summary":"Google Cloud는 Always-On Memory Agent를 발표했습니다. 이는 ADK와 Gemini 3.1 Flash-Lite를 기반으로 하는 24/7 백그라운드 프로세스로, 벡터 데이터베이스나 임베딩 벡터를 사용하지 않고 LLM이 구조화된 메모리를 SQLite에 기록합니다. 이 솔루션은 30분마다 메모리를 자동으로 통합하고 연관성을 발견하며, 27가지 파일 형식을 지원하고 HTTP API 또는 Streamlit 대시보드를 통해 상호작용할 수 있습니다.","category":"제품 업데이트","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud는 Always-On Memory Agent를 출시하여 LLM으로 RAG 및 벡터 데이터베이스를 지속적으로 통합합니다 - Aioga AI 뉴스","description":"Google Cloud는 Always-On Memory Agent를 발표했습니다. 이는 ADK와 Gemini 3.1 Flash-Lite를 기반으로 하는 24/7 백그라운드 프로세스로, 벡터 데이터베이스나 임베딩 벡터를 사용하지 않고 LLM이 구조화된 메모리를 SQLite에 기록합니다. 이 솔루션은 30분마다 메모리를 자동...","url":"https://www.aioga.com/ko/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:03:20.867Z"},"es":{"title":"Google Cloud lanza Always-On Memory Agent, utilizando LLM para integrar continuamente como alternativa a RAG y bases de datos vectoriales","summary":"Google Cloud lanza Always-On Memory Agent, un proceso en segundo plano 24/7 basado en ADK y Gemini 3.1 Flash-Lite, que no utiliza bases de datos vectoriales ni vectores incrustados, sino que LLM escribe la memoria estructurada en SQLite. Esta solución integra automáticamente la memoria y descubre asociaciones cada 30 minutos, soporta 27 tipos de archivos, e interactúa a través de API HTTP o el panel de control Streamlit.","category":"Productos","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud lanza Always-On Memory Agent, utilizando LLM para integrar continuamente como alternativa a RAG y bases de datos vectoriales - Aioga Noticias de IA","description":"Google Cloud lanza Always-On Memory Agent, un proceso en segundo plano 24/7 basado en ADK y Gemini 3.1 Flash-Lite, que no utiliza bases de datos vectoriales ni vectores incrustados...","url":"https://www.aioga.com/es/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:03:20.891Z"},"fr":{"title":"Google Cloud lance Always-On Memory Agent, utilisant LLM pour intégrer en continu en remplacement de RAG et des bases de données vectorielles","summary":"Google Cloud a lancé Always-On Memory Agent, un processus en arrière-plan 24/7 basé sur ADK et Gemini 3.1 Flash-Lite, qui n’utilise ni base de données vectorielle ni vecteurs d’intégration, mais écrit la mémoire structurée dans SQLite via LLM. Ce système intègre automatiquement la mémoire et découvre les corrélations toutes les 30 minutes, prend en charge 27 types de fichiers, et permet l’interaction via une API HTTP ou un tableau de bord Streamlit.","category":"Produits","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud lance Always-On Memory Agent, utilisant LLM pour intégrer en continu en remplacement de RAG et des bases de données vectorielles - Aioga Actualités IA","description":"Google Cloud a lancé Always-On Memory Agent, un processus en arrière-plan 24/7 basé sur ADK et Gemini 3.1 Flash-Lite, qui n’utilise ni base de données vectorielle ni vecteurs d’int...","url":"https://www.aioga.com/fr/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:04:04.749Z"},"de":{"title":"Google Cloud führt den Always-On Memory Agent ein, der LLM zur kontinuierlichen Integration als Ersatz für RAG und Vektordatenbanken verwendet","summary":"Google Cloud hat den Always-On Memory Agent veröffentlicht, einen 24/7-Hintergrundprozess basierend auf ADK und Gemini 3.1 Flash-Lite, der keine Vektordatenbank oder eingebetteten Vektor verwendet, sondern dass LLM strukturierte Erinnerungen in SQLite schreibt. Die Lösung integriert Erinnerungen und erkennt Zusammenhänge alle 30 Minuten automatisch, unterstützt 27 Dateitypen und ermöglicht Interaktion über HTTP-API oder Streamlit-Dashboard.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud führt den Always-On Memory Agent ein, der LLM zur kontinuierlichen Integration als Ersatz für RAG und Vektordatenbanken verwendet - Aioga KI-News","description":"Google Cloud hat den Always-On Memory Agent veröffentlicht, einen 24/7-Hintergrundprozess basierend auf ADK und Gemini 3.1 Flash-Lite, der keine Vektordatenbank oder eingebetteten...","url":"https://www.aioga.com/de/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:04:05.070Z"},"pt-BR":{"title":"O Google Cloud lança o Always-On Memory Agent, usando LLM para integração contínua como alternativa ao RAG e bancos de dados vetoriais","summary":"O Google Cloud lançou o Always-On Memory Agent, um processo em segundo plano 24/7 baseado em ADK e Gemini 3.1 Flash-Lite, que não utiliza banco de dados vetorial ou vetores incorporados, mas sim um LLM que escreve memórias estruturadas no SQLite. Esta solução integra memórias automaticamente a cada 30 minutos e descobre correlações, suporta 27 tipos de arquivos e permite interação através de API HTTP ou painel Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"O Google Cloud lança o Always-On Memory Agent, usando LLM para integração contínua como alternativa ao RAG e bancos de dados vetoriais - Aioga Notícias de IA","description":"O Google Cloud lançou o Always-On Memory Agent, um processo em segundo plano 24/7 baseado em ADK e Gemini 3.1 Flash-Lite, que não utiliza banco de dados vetorial ou vetores incorpo...","url":"https://www.aioga.com/pt-BR/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:04:56.439Z"},"ru":{"title":"Google Cloud запустил Always-On Memory Agent, который с помощью LLM непрерывно интегрируется, заменяя RAG и векторные базы данных","summary":"Google Cloud выпустил Always-On Memory Agent, это фоновый процесс, работающий 24/7 на основе ADK и Gemini 3.1 Flash-Lite, который не использует векторные базы данных или векторные встраивания, а записывает структурированную память в SQLite с помощью LLM. Решение автоматически интегрирует память и обнаруживает связи каждые 30 минут, поддерживает 27 типов файлов и взаимодействует через HTTP API или панель Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud запустил Always-On Memory Agent, который с помощью LLM непрерывно интегрируется, заменяя RAG и векторные базы данных - Aioga Новости ИИ","description":"Google Cloud выпустил Always-On Memory Agent, это фоновый процесс, работающий 24/7 на основе ADK и Gemini 3.1 Flash-Lite, который не использует векторные базы данных или векторные...","url":"https://www.aioga.com/ru/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:04:54.401Z"},"ar":{"title":"أطلقت Google Cloud وكيل الذاكرة Always-On، باستخدام LLM للاندماج المستمر كبديل عن RAG وقواعد بيانات المتجهات","summary":"أطلقت Google Cloud أداة Always-On Memory Agent، وهي عملية خلفية تعمل على مدار الساعة تعتمد على ADK و Gemini 3.1 Flash-Lite، ولا تستخدم قاعدة بيانات متجهية أو تضمين المتجهات، بل يقوم LLM بكتابة الذاكرة المهيكلة في SQLite. تقوم هذه الحلول بدمج الذاكرة واكتشاف العلاقات تلقائيًا كل 30 دقيقة، وتدعم 27 نوعًا من الملفات، ويمكن التفاعل معها عبر HTTP API أو لوحة Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"أطلقت Google Cloud وكيل الذاكرة Always-On، باستخدام LLM للاندماج المستمر كبديل عن RAG وقواعد بيانات المتجهات - Aioga أخبار الذكاء الاصطناعي","description":"أطلقت Google Cloud أداة Always-On Memory Agent، وهي عملية خلفية تعمل على مدار الساعة تعتمد على ADK و Gemini 3.1 Flash-Lite، ولا تستخدم قاعدة بيانات متجهية أو تضمين المتجهات، بل يقو...","url":"https://www.aioga.com/ar/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:05:41.651Z"},"hi":{"title":"Google Cloud ने Always-On Memory Agent लॉन्च किया, LLM का उपयोग करके RAG और वेक्टर डेटाबेस के विकल्प के रूप में लगातार एकीकृत किया","summary":"Google Cloud ने Always-On Memory Agent जारी किया, एक 24/7 बैकग्राउंड प्रक्रिया जो ADK और Gemini 3.1 Flash-Lite पर आधारित है, जो वेक्टर डेटाबेस या एम्बेडेड वेक्टर का उपयोग नहीं करती, बल्कि LLM द्वारा संरचित मेमोरी को SQLite में लिखा जाता है। यह समाधान हर 30 मिनट में स्वचालित रूप से मेमोरी को संयोजित करता है और संबंधों की खोज करता है, 27 प्रकार की फ़ाइलों का समर्थन करता है, और HTTP API या Streamlit डैशबोर्ड के माध्यम से इंटरैक्ट करता है।","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud ने Always-On Memory Agent लॉन्च किया, LLM का उपयोग करके RAG और वेक्टर डेटाबेस के विकल्प के रूप में लगातार एकीकृत किया - Aioga AI समाचार","description":"Google Cloud ने Always-On Memory Agent जारी किया, एक 24/7 बैकग्राउंड प्रक्रिया जो ADK और Gemini 3.1 Flash-Lite पर आधारित है, जो वेक्टर डेटाबेस या एम्बेडेड वेक्टर का उपयोग नहीं करती...","url":"https://www.aioga.com/hi/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:05:54.890Z"},"it":{"title":"Google Cloud lancia Always-On Memory Agent, utilizza LLM per integrare continuamente come alternativa a RAG e ai database vettoriali","summary":"Google Cloud ha lanciato Always-On Memory Agent, un processo in background 24/7 basato su ADK e Gemini 3.1 Flash-Lite, che non utilizza database vettoriali o vettori incorporati, ma lascia che LLM scriva la memoria strutturata in SQLite. Questa soluzione integra automaticamente la memoria e scopre le correlazioni ogni 30 minuti, supporta 27 tipi di file, e permette l'interazione tramite API HTTP o dashboard Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud lancia Always-On Memory Agent, utilizza LLM per integrare continuamente come alternativa a RAG e ai database vettoriali - Aioga Notizie IA","description":"Google Cloud ha lanciato Always-On Memory Agent, un processo in background 24/7 basato su ADK e Gemini 3.1 Flash-Lite, che non utilizza database vettoriali o vettori incorporati, m...","url":"https://www.aioga.com/it/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:06:37.958Z"},"nl":{"title":"Google Cloud introduceert Always-On Memory Agent, gebruikt LLM voor continue integratie ter vervanging van RAG en vector databases","summary":"Google Cloud heeft de Always-On Memory Agent gelanceerd, een 24/7 achtergrondproces gebaseerd op ADK en Gemini 3.1 Flash-Lite, dat geen vectordatabase of embedded vectors gebruikt, maar waarin LLM gestructureerde herinneringen naar SQLite schrijft. Het systeem integreert automatisch herinneringen en ontdekt verbanden elke 30 minuten, ondersteunt 27 bestandstypen, en kan worden bediend via een HTTP API of Streamlit-dashboard.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud introduceert Always-On Memory Agent, gebruikt LLM voor continue integratie ter vervanging van RAG en vector databases - Aioga AI-nieuws","description":"Google Cloud heeft de Always-On Memory Agent gelanceerd, een 24/7 achtergrondproces gebaseerd op ADK en Gemini 3.1 Flash-Lite, dat geen vectordatabase of embedded vectors gebruikt,...","url":"https://www.aioga.com/nl/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:06:34.811Z"},"tr":{"title":"Google Cloud, RAG ve vektör veritabanının yerine sürekli entegrasyon için LLM kullanan Always-On Memory Agent'ı tanıttı","summary":"Google Cloud, ADK ve Gemini 3.1 Flash-Lite tabanlı, vektör veritabanı veya yerleştirilmiş vektör kullanmayan ve yapay zekânın yapılandırılmış hafızayı SQLite'a yazdığı 7/24 arka plan süreci Always-On Memory Agent'i yayınladı. Bu çözüm, hafızayı her 30 dakikada bir otomatik olarak bütünleştirir ve ilişkileri keşfeder, 27 dosya türünü destekler ve HTTP API veya Streamlit panosu üzerinden etkileşim sağlar.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud, RAG ve vektör veritabanının yerine sürekli entegrasyon için LLM kullanan Always-On Memory Agent'ı tanıttı - Aioga AI Haberleri","description":"Google Cloud, ADK ve Gemini 3.1 Flash-Lite tabanlı, vektör veritabanı veya yerleştirilmiş vektör kullanmayan ve yapay zekânın yapılandırılmış hafızayı SQLite'a yazdığı 7/24 arka pl...","url":"https://www.aioga.com/tr/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:07:24.756Z"},"vi":{"title":"Google Cloud ra mắt Always-On Memory Agent, sử dụng LLM liên tục tích hợp để thay thế RAG và cơ sở dữ liệu vector","summary":"Google Cloud ra mắt Always-On Memory Agent, một tiến trình nền 24/7 dựa trên ADK và Gemini 3.1 Flash-Lite, không sử dụng cơ sở dữ liệu vector hoặc embedding vector, mà được LLM ghi nhớ cấu trúc vào SQLite. Giải pháp này tự động tổng hợp ký ức và phát hiện các mối liên hệ mỗi 30 phút, hỗ trợ 27 loại tệp, tương tác qua HTTP API hoặc bảng điều khiển Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud ra mắt Always-On Memory Agent, sử dụng LLM liên tục tích hợp để thay thế RAG và cơ sở dữ liệu vector - Tin tức AI Aioga","description":"Google Cloud ra mắt Always-On Memory Agent, một tiến trình nền 24/7 dựa trên ADK và Gemini 3.1 Flash-Lite, không sử dụng cơ sở dữ liệu vector hoặc embedding vector, mà được LLM ghi...","url":"https://www.aioga.com/vi/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:07:29.113Z"},"id":{"title":"Google Cloud meluncurkan Always-On Memory Agent, menggunakan LLM untuk terus mengintegrasikan sebagai pengganti RAG dan basis data vektor","summary":"Google Cloud merilis Always-On Memory Agent, sebuah proses latar belakang 24/7 yang berbasis ADK dan Gemini 3.1 Flash-Lite, tidak menggunakan database vektor atau vektor embedding, melainkan oleh LLM yang menulis memori terstruktur ke SQLite. Solusi ini secara otomatis mengintegrasikan memori dan menemukan hubungan setiap 30 menit, mendukung 27 jenis file, dan dapat berinteraksi melalui HTTP API atau dashboard Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud meluncurkan Always-On Memory Agent, menggunakan LLM untuk terus mengintegrasikan sebagai pengganti RAG dan basis data vektor - Berita AI Aioga","description":"Google Cloud merilis Always-On Memory Agent, sebuah proses latar belakang 24/7 yang berbasis ADK dan Gemini 3.1 Flash-Lite, tidak menggunakan database vektor atau vektor embedding,...","url":"https://www.aioga.com/id/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:08:09.781Z"},"th":{"title":"Google Cloud เปิดตัว Always-On Memory Agent ใช้ LLM เพื่อรวมอย่างต่อเนื่องแทน RAG และฐานข้อมูลเวกเตอร์","summary":"Google Cloud เปิดตัว Always-On Memory Agent กระบวนการแบ็คกราวด์ 24/7 ที่สร้างขึ้นบน ADK และ Gemini 3.1 Flash-Lite โดยไม่ใช้ฐานข้อมูลเวกเตอร์หรือเวกเตอร์ฝัง แต่ใช้ LLM ในการเขียนความจำเชิงโครงสร้างลงใน SQLite โซลูชันนี้จะรวมความจำและค้นหาความสัมพันธ์โดยอัตโนมัติทุก 30 นาที รองรับไฟล์ 27 ประเภท และสามารถโต้ตอบผ่าน HTTP API หรือแดชบอร์ดของ Streamlit","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud เปิดตัว Always-On Memory Agent ใช้ LLM เพื่อรวมอย่างต่อเนื่องแทน RAG และฐานข้อมูลเวกเตอร์ - ข่าว AI Aioga","description":"Google Cloud เปิดตัว Always-On Memory Agent กระบวนการแบ็คกราวด์ 24/7 ที่สร้างขึ้นบน ADK และ Gemini 3.1 Flash-Lite โดยไม่ใช้ฐานข้อมูลเวกเตอร์หรือเวกเตอร์ฝัง แต่ใช้ LLM ในการเขียนควา...","url":"https://www.aioga.com/th/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:08:22.145Z"},"pl":{"title":"Google Cloud wprowadza Always-On Memory Agent, wykorzystujący LLM do ciągłej integracji, zastępując RAG i bazy danych wektorowych","summary":"Google Cloud wprowadził Always-On Memory Agent, 24/7 proces działający w tle oparty na ADK i Gemini 3.1 Flash-Lite, który nie korzysta z baz danych wektorowych ani osadzonych wektorów, lecz zapisuje ustrukturyzowaną pamięć za pomocą LLM do SQLite. Rozwiązanie automatycznie integruje pamięć i odkrywa powiązania co 30 minut, obsługuje 27 typów plików i umożliwia interakcję przez HTTP API lub pulpit Streamlit.","category":"产品更新","source":"MarkTechPost（RSS）","aggregationSource":"MarkTechPost（RSS）","pageTitle":"Google Cloud wprowadza Always-On Memory Agent, wykorzystujący LLM do ciągłej integracji, zastępując RAG i bazy danych wektorowych - Aioga Wiadomości AI","description":"Google Cloud wprowadził Always-On Memory Agent, 24/7 proces działający w tle oparty na ADK i Gemini 3.1 Flash-Lite, który nie korzysta z baz danych wektorowych ani osadzonych wekto...","url":"https://www.aioga.com/pl/news/cmrq3c6fz0083bi6vbya5443p/","contentTranslated":true,"sourceHash":"927c69f0a29b8100","translatedAt":"2026-07-22T22:09:10.144Z"}}}}