AI NEWSLatest AI NewsArticleTikTok Upgrades Minors’ Mode Engine, Brings Multimodal Large Language Model into Age-Appropriate RecommendationsPublished in Latest AI NewsTime :Jul 27, 2026Read :4minuteJuly 27th news: Today, the Douyin Group issued a notice stating that in order to alleviate parents’ concerns about supervision and respond to the growth needs of minors, the platform has comprehensively upgraded the core engine of the minor mode - the age-appropriate recommendation algorithm. Once users activate the minor mode and fill in their birth date, this specialized age-appropriate recommendation algorithm will be activated.This round of comprehensive upgrade of the minor mode recommendation algorithm focuses on introducing cutting-edge multimodal large language model (MLLM) technology, thus building an intelligent and refined content identification and distribution system. Multimodal large language models inherently possess cross-modal semantic understanding capabilities, capable of processing text, images, audio, and other types of information simultaneously, while also considering dimensions such as cognitive complexity and value orientation of the content, determining the appropriate age range for each video, and laying a technical foundation for subsequent age-appropriate recommendations.The process works in several steps: First, the platform uses a high-quality multimodal large model for minors to screen a massive amount of information, selecting candidate content suitable for minors to watch; then it is handed over to human experts for strict review, selecting high-quality content with positive values, and including them in the exclusive content pool for minors; subsequently, the age-identification multimodal large model deeply understands and classifies these reviewed high-quality contents, outputting the appropriate age range information.The official also mentioned that the platform currently adopts a detailed age classification standard for age-appropriate recommendations of high-quality content. This mechanism can adapt to the cognitive characteristics and psychological changes of minors at different stages, making the granularity of content supply more delicate; meanwhile, the algorithm will also recommend content based on the child’s preferences. This “human screening + large model classification + age-appropriate recommendation + content preference matching” multi-layered guarantee, within the scope of the information the platform can collect, tries its best to make each recommendation more in line with the child’s cognitive development needs; children can also give real-time feedback by long-pressing and clicking the “Not interested” button.

AI NEWSLatest AI NewsArticleGoogle AI Search Doubles in a Year: 43% of Search Results Now Have AI Overview, Traditional Web Pages Are Being Replaced by Direct AnswersPublished in Latest AI NewsTime :Jul 28, 2026Read :5minuteSimilarweb’s latest report reveals an undeniable trend: AI search is rapidly reshaping traditional search experiences. The presence of Google’s AI Overview in search results has increased from 15% a year ago to 43%, and AI-driven information retrieval is quietly becoming the default user experience.Data shows that traffic to Google’s AI mode rose from 126 million in June 2025 to 279 million in May 2026, more than doubling. At the same time, user search behavior is also undergoing a fundamental change — more people are beginning to use longer, more natural conversational queries, replacing the short keyword-based searches of the past. Searching is shifting from “helping users find web pages” to “providing answers directly to users.“References have increased fivefold, but traffic does not necessarily followThis change has profoundly affected the logic of internet traffic distribution. Traditional publishers and content websites are facing the dilemma of AI summaries “siphoning” traffic — users may get answers directly on the search results page without even clicking into the original web page.Data shows that the number of search responses containing AI references has grown more than fivefold over the past year, but references do not equal traffic. For example, in the U.S. market, only 6.8% of desktop search queries on ChatGPT as of May 2026 included reference links. However, there are significant differences across industries — travel, retail, and sports sectors are more likely to gain AI search references, and these areas benefit more significantly in terms of traffic.For content creators, it’s not all bad news. After ChatGPT’s search update on May 7, 2026, the web recommendation traffic from ChatGPT to U.S. desktop users improved significantly — by the end of May, the percentage of web visits increased from 25% to nearly 60%. This means that when AI products begin to prioritize references and traffic redirection, the content ecosystem still has room for recovery.From a navigation tool to a consumption platform, Google’s identity has changedIn general, AI search is driving a structural transformation in internet access points. Google has shifted from a traditional “web navigation tool” to an “information consumption platform,” and the search results page is no longer just a list of links, but a destination for direct information delivery.For websites and content organizations, the rules of the game have been rewritten. Optimizing content structure and AI visibility in an AI-driven information distribution model will determine the next phase of the content competition landscape. The traditional model that heavily relied on search traffic is being slowly dismantled by an invisible revolution.

AI NEWSLatest AI NewsArticleChina’s On-Device Large Model Makes Its Debut on Global Flagship Devices: MiniCPM by Mingshi is Integrated into Samsung Galaxy Z Fold8 SeriesPublished in Latest AI NewsTime :Jul 23, 2026Read :3minuteSamsung unveiled the Galaxy Z Fold8, Galaxy Z Fold8 Ultra, and Galaxy Z Flip8 at the Galaxy Unpacked event in London last night. Just today, MinaCPM, a series of edge-side large models from Minding Intelligence, has deeply empowered Samsung Galaxy AI - this is the first time a domestic edge-side large model has entered the global flagship product line of an international top smartphone manufacturer.According to the introduction, this time Minding Intelligence provides Samsung Galaxy AI with edge-side model capabilities, covering text understanding and multimodal perception. The new devices that have truly adopted this “domestic brain” are the Galaxy Z Fold8 and Galaxy Z Fold8 Ultra.These two devices also have their own strengths. The Galaxy Z Fold8 features a new 4:3 ratio “wide fold” design, with an internal screen of 8 inches, dual 50-megapixel rear cameras, and a 4800mAh battery, completely moving away from the traditional narrow ratio of foldable screens, offering a visual and office experience closer to a tablet. The Galaxy Z Fold8 Ultra is the first Ultra model in Samsung’s foldable family, continuing the book-like fold design, equipped with a 200-megapixel main camera and a 5000mAh battery, firmly securing its position as the top of the foldable flagship segment.

AI NEWSLatest AI NewsArticleByteDance Restructures Its AI Business Landscape: Large Model ARR Reaches $4 Billion, Integrated with Collaboration and Core Large ModelsPublished in Latest AI NewsTime :Jul 30, 2026Read :2minuteOn July 30, ByteDance announced a major restructuring of its artificial intelligence business organization, splitting the team behind its collaboration office application, Feishu, into two parts, which were respectively integrated into the Doubao and Volcano Engine teams. This move aims to further strengthen the synergy between large model products and cloud services in enterprise productivity scenarios.According to the information, the core of this organizational adjustment lies in breaking down departmental barriers. The Feishu product team will be officially integrated with the Doubao product team to form a new Doubao product team, which will be comprehensively managed by Zhao Qi, the head of Doubao. Feishu’s original leader, Xie Xin, will report to Zhao Qi. At the same time, the remaining team members have been connected with Volcano Engine, accelerating commercialization and technological empowerment.In terms of business growth, ByteDance revealed its latest financial performance. According to the average consumption calculation for July this year, the annualized revenue run rate of its large model business has reached 4 billion US dollars, a figure that surpasses the total of other domestic large model companies, demonstrating strong market momentum and commercialization capabilities.

AI NEWSLatest AI NewsArticleWorkBuddy Launches ‘Human-AI Dual Writing’: AI Exits the Chat Window and Joins Your Document to Work TogetherPublished in Latest AI NewsTime :Jul 30, 2026Read :4minuteOn July 30, the office AI platform WorkBuddy officially launched a major update - the “Human-AI Dual Writing” feature, which for the first time enables real-time collaboration between AI and users within the same document interface, marking the transition of AI office work from “add-on assistance” to the “synchronized collaboration” stage.Differing from traditional AI assistants that require repeated copy-pasting and window switching, “Human-AI Dual Writing” allows AI to directly enter office scenarios such as documents, spreadsheets, and PPTs. Users only need to select content and input natural language instructions, and AI can complete modifications, formatting, and analysis directly in the original file, with all changes taking effect immediately. This feature is the first of its kind in the industry, supporting both Tencent Online Documents and local Office files (Word, Excel, PPT, MD), truly achieving a seamless experience of “edit whatever you think of.“In practical office scenarios, this feature demonstrates significant efficiency advantages. In document scenarios, users can generate weekly summaries, batch adjust full-text formatting, correct typos, and refine content with just one sentence; in spreadsheet scenarios, AI can automatically generate pivot tables, fill formulas, clean data, mark abnormal values, and output visual charts and analysis reports; in PPT scenarios, it supports unified style, automatic image insertion, page structure splitting, core idea summarization, and voiceover notes generation, significantly shortening the production cycle. According to actual test cases, technical proposal writing was reduced from one day to just a few minutes, market research from one week to a morning, and bid document processing from 3-5 days to half a day, with an overall efficiency improvement of over 10 times.With the release of this feature, WorkBuddy also launched the “Share Your Usage and Win Prizes” campaign, inviting users to experience “Human-AI Dual Writing” and share real case studies, with the top prize being 2000 points and a Tencent Document seasonal card. The campaign will continue until August 16, 2026.

AI NEWSLatest AI NewsArticleNadella Opens Up: Don’t Just Trust OpenAI and Anthropic, Microsoft Will Be the CheapestPublished in Latest AI NewsTime :Jul 30, 2026Read :4minuteMicrosoft is in an unprecedented delicate position - both one of the world’s largest cloud service providers and holding significant shares in two leading AI labs, OpenAI and Anthropic. However, as both companies continue to expand into the application layer and intelligent agent infrastructure, trying to directly control customer relationships, CEO Satya Nadella clearly isn’t going to let this trend threaten Microsoft’s cash cow.Microsoft has just released a stunning quarterly report: revenue of $90 billion and net profit of $35.8 billion for the fourth quarter, and total revenue of $331.8 billion and net profit of $133.7 billion for the 2026 fiscal year. Nadella’s strategy is clear - he has repeatedly warned enterprise customers recently not to rely solely on cutting-edge AI labs’ models, as binding intelligent agent frameworks with underlying models is dangerous because it means handing over a large amount of internal secrets to model vendors whose credibility is questionable.From the Hugging Face incident to self-developed MAI, Microsoft reveals its entire handIn the Wednesday earnings call, Nadella directly laid out his cards to Wall Street analysts: this is Microsoft’s opportunity. He outlined the core architectural principle of “separating intelligent agent frameworks from models,” ensuring that any model can be replaced at any time - and Microsoft sells a range of intelligent agent frameworks under the Copilot brand, including GitHub Copilot. He even used last week’s Hugging Face intrusion incident as evidence: after an unannounced model from OpenAI broke through the sandbox and invaded Hugging Face, the latter initially tried to use a private cutting-edge model to analyze the attack, but the model refused to help, and ultimately had to turn to the Chinese open-source model Z.ai GLM 5.2 to complete the defense.Nadella also revealed Microsoft’s self-developed arsenal: the MAI series models paired with self-developed Maya chips, which achieve a 40% performance improvement per watt when running on Maya 200. Microsoft Cloud offers more than 11,000 models, covering OpenAI, Anthropic, Mistral, xAI, and the self-developed MAI family. The newly released MAI Cyber One Flash claims to surpass the larger Mythos model from Anthropic at half the cost. Nadella’s message is clear enough: cutting-edge models can be used, but shouldn’t be trusted to the point of dependence - and Microsoft will be the option that allows you to switch anytime and affordably.

AI NEWSLatest AI NewsArticleMicrosoft Confirms Development of Copilot AI Super App, Integrating Chat, Coding, and Intelligent Agent CapabilitiesPublished in Latest AI NewsTime :Jul 30, 2026Read :3minuteMicrosoft is developing an AI “super app” that integrates Copilot chat, coding assistant, and intelligent agent capabilities. Satya Nadella, Microsoft’s CEO, confirmed in the latest earnings call that the app is planned to launch this year and will cover both consumer and business use cases.Nadella stated that Copilot is rapidly evolving from a simple chat assistant to areas such as collaborative work and automation. Microsoft will integrate different Copilot experiences this quarter, including code capabilities, and incorporate them into a unified super app platform. He believes this will be an important step in Copilot’s development and more details will be announced in the future.This confirmation responds to previous reports by Fortune magazine about Microsoft’s development of an AI super app. According to the report, the app will integrate multiple Microsoft AI product capabilities, including Copilot chatbot for general interactions, GitHub Copilot for developers, Copilot Cowork for team collaboration, and an intelligent agent system with task automation capabilities.Microsoft’s move also reflects the shift in AI assistant competition from single-function tools to comprehensive AI platforms. Previously, OpenAI also launched a new ChatGPT Work app, combining ChatGPT with the AI coding tool Codex, exploring an integrated AI experience in office scenarios.In terms of finance, Microsoft’s revenue in the last quarter reached $90 billion, with artificial intelligence and cloud services being key drivers of growth. As major tech companies continue to integrate models, tools, and agent capabilities, AI super apps may become an important gateway for both enterprise and individual users to access artificial intelligence in the future.

ENLatest AI NewsMCP Startup Runlayer Accuses Rippling of Allegedly Stealing Its Product IdeasAI Startup Runlayer, with its MCP Security Gateway as the core, accuses the HR software company Rippling of stealing its ideas and trade secrets after a year of deep collaboration and product trial, and has officially filed a lawsuit. This dispute highlights the trust crisis in enterprise procurement within the AI infrastructure sector, as well as the intense conflict between large companies and startups regarding technology collaboration and trade secret boundaries.just now7.4KBuilding a Learning AI Customer Service: Encore AI Secures $30 Million in Series A Funding to Monetize Enterprise Conversation DataEncore AI has secured $30 million in Series A funding, led by Team8 with participation from Planven and other institutions. The company specializes in analyzing enterprise customer conversations, training and deploying intelligent voice agents that can operate independently or assist teams. Its predecessor was founded in 2022 and previously provided recommendation software for financial advisors; it has since rebranded and shifted its core business to deploying voice agents driven by conversation analysis.just now5.6KMicrosoft Confirms Development of Copilot AI Super App, Integrating Chat, Coding, and Intelligent Agent CapabilitiesMicrosoft is developing an AI super app that integrates Copilot chat, coding assistant, and intelligent agent capabilities, with plans to launch it this year, covering both consumer and business scenarios. CEO Nadella stated that Copilot is shifting from chat to collaborative office work and automation, and the company will unify different Copilot experiences this quarter.just now8.7KNadella Opens Up: Don’t Just Trust OpenAI and Anthropic, Microsoft Will Be the CheapestMicrosoft, a top cloud provider, invests in OpenAI and Anthropic. But they’re expanding into application layer and agent infrastructure…..just now6.8KWorkBuddy Launches ‘Human-AI Dual Writing’: AI Exits the Chat Window and Joins Your Document to Work TogetherWorkBuddy introduces “Human-AI Co-writing”, enabling real-time collaborative editing within the same document interface. Users select content and use natural language instructions for AI to modify or format directly in the original file, eliminating copy-paste or window switching. This marks a shift from external AI assistance to in-screen collaboration…..just now9.1KByteDance Restructures Its AI Business Landscape: Large Model ARR Reaches $4 Billion, Integrated with Collaboration and Core Large ModelsOn July 30, ByteDance restructured its AI division: splitting the Feishu team, integrating the product team with Doubao under Zhao Qi, and merging the rest into Volcano Engine. The move aims to break down silos and strengthen synergy between large models and cloud services in enterprise productivity…..just now6.7KZuckerberg Claims: Billions of People Will Have Personal AI Assistants Within Five Years, Meta Invests Heavily in the FutureZuckerberg sees intelligent agents integrating into daily finance, health, social, and home life. As multi-agent interactions grow, messaging platforms like WhatsApp become key AI hubs. Tech giants are racing to build agentic systems that act on users’ behalf…..just now7.1KFeyn Team Launches New Background Removal Open-Source Model FeyNoBgThe Feyn team has introduced the cutting-edge background removal model FeyNoBg and open-sourced the NoBg training and inference library. Based on the BiRefNet architecture with deep optimization, it enhances the key feature extraction module, improving the precise handling of edge details such as hair and sheer fabrics, providing a lightweight and efficient segmentation solution.just now18.2KVolcano Engine Officially Launches Doubao Search Open Service to Empower Intelligent Agent Long-Term TasksVolcano Engine launched Doubao Search Service to support AI agents in long-task deep retrieval. Designed for enterprises and developers, it enables cross-language, multimodal, and multi-vertical web data access, enhances information credibility governance, and outputs structured data and Markdown for precise queries…..just now8.8KxAI releases Grok Voice Think Fast 2.0 with comprehensive upgrades in speech recognition and AI agent capabilitiesxAI releases the Grok Voice Think Fast 2.0 speech recognition model for developers, achieving improvements in intelligent level, transcription accuracy, conversation ability, and tool calling efficiency. The pricing is $0.08 per minute. This model enhances performance without requiring adjustments to existing prompts. It achieved a total score of 82.9% in the Artificial Analysis benchmark test, surpassing the previous version’s 75.7%.just now10.9KTencent Open Sources AngelSpec Framework: Solving the Real Inference Efficiency Challenges of Large ModelsTencent open-sources the unified training framework AngelSpec, which reduces the cost of autoregressive decoding for lar

Seko是什么Seko是商汤科技推出的商汤科技推出的全球首个创编一体的AI视频创作Agent，基于Seedance 2.0多模态大模型，用户通过自然语言对话输入创意，Agent自动完成剧本拆解、分镜规划、画面生成、配音配乐到成片导出全流程。Seko 3.0版本全量上线无限画布，支持AI自动生成画布工作流、多剧集资产贯通与影视级工具箱，实现从灵感到精细化生产的完整闭环。Seko的主要功能全流程Agent创作：输入创意自动完成剧本→分镜→画面→视频→配音一站式成片。无限画布工作流：AI自动生成包含角色、场景、道具、分镜的完整画布，节点式关联参考图与提示词。多剧集连续生成：支持100集以内剧本连贯创作，跨集相同资产智能调度保持统一。角色一致性引擎：自研SekoIDX技术确保同一角色跨集、跨分镜形象稳定，避免”换脸”穿帮。多人对口型：SekoTalk技术支持同一画面内3人以上同时对话，口型精准匹配。多模型自由切换：集成Seedance 2.0、可灵3.0 Omni、Vidu、Wan等主流视频模型。画布精修工具：提供消除笔、局部重绘、元素添加、文字修改等分镜级图像编辑。影视级工具箱：内置720°全景、多角度、九宫格、故事推演等专业影像工具。自然语言编辑：对话式增删分镜、修改角色画风、替换背景音乐，所见即改。精选爆款Skill：平台精选顶尖爆款实用Skill，帮助用户轻松复刻热门内容风格。如何使用Seko访问平台：访问Seko官网https://seko.sensetime.com，注册和登录。灵感输入：在首页输入框输入创意灵感，或上传本地剧本文件；开启多剧集开关可一次性创建多集内容。模型与模式选择：选择生图模型，选定画面比例，并选择 Seedance 2.0 全能模式或短剧/音乐 MV/知识分享等入口。策划案确认：AI 自动生成策划案后，检查摘要、美术风格、角色主体、场景概念与分镜剧本，通过对话或手动编辑修改内容。分镜图调整：点击生成视频获得分镜图，对不满意的分镜进行替换、改图、画布编辑或提示词优化。视频生成：选择视频模型与分辨率，将分镜图转为视频；支持首尾帧转场、一键转视频及对口型。配音配乐：为分镜添加旁白配音和背景音乐，调整音轨位置与时长。成片导出：点击导出视频，选择有水印或无水印，导出完整视频或独立分镜文件。Seko的核心优势Seedance 2.0 视频大模型：基于多模态视频大模型，融合文生图、图生视频、对口型等能力。Agent 自动拆解架构：AI 自动将剧本拆解为分镜表，通过主体库锁定角色与场景特征，确保跨分镜一致性。多模型集成调度：集成 Seedance、可灵、Vidu、Wan、Pixverse、海螺等第三方视频模型，编辑器内可随时切换。AI 润色提示词引擎：自动将一句话描述扩展为完整提示词（画风+景别+主体+背景），降低专业门槛。Seko的同类竞品对比对比维度SekoPexoAIpai产品定位一站式 AI 视频创作平台，剧本到成片全流程 Agent 驱动AI 视频创作 Agent，定位为”个人 AI 视频伙伴”，零门槛对话式创作智能 AI 视频生成工具，主打”氛围感视频”（Vibe Videoing）创作核心交互结构化工作流：灵感输入 → 策划案 → 分镜图 → 视频 → 导出单聊天框对话式交互，像发微信一样描述需求，AI 主动规划并交付模板驱动 + “AIpai it” 创意入口，输入想法或上传素材一键生成工作流深度深度可控，支持分镜级精修、画布编辑、提示词优化、音轨调整极简自动化，AI 自动完成脚本、分镜、配音、转场、BGM，用户仅验收中等自动化，内置 Agent 优化剪辑/配乐/转场，支持模板快速套用剧本分镜能力支持上传剧本自动拆解分镜表，Agent 生成完整策划案，分镜级逐帧控制支持自然语言描述后 AI 自动构建故事板，但不支持上传完整剧本拆解支持输入创意想法生成视频，无原生剧本到分镜拆解能力角色一致性主体库锁定角色特征，支持跨分镜长期一致性保持，适合连续剧通过参考图保持单片段一致性，跨片段连续性依赖 AI 自动处理模板风格统一，角色一致性主要依赖素材和模板匹配对口型能力支持单人/多人（至多 3 人同框），单图对口型最长 60 秒支持中文口播与口型同步，具体人数与时长限制未明确未明确强调对口型功能，侧重氛围感与音乐节奏匹配模型策略集成 Seedance、可灵、Vidu、Wan、Pixverse、海螺等多模型，编辑器内可切换智能路由 Seedance、Sora、Kling 等顶尖模型，用户无需手动选择未公开底层模型细节，内置 AI 模型自动处理画布精修提供消除笔、局部重绘、元素添加、文字修改等完整画布编辑工具支持时间轴标记精准修改，圈出区域提出调整需求，但无专业画布编辑未明确提供画布级精修功能，侧重一键成片多剧集管理原生支持多剧集创建、自动拆解、逐集上传与连续项目管理不支持原生多剧集连载管理，以单条视频创作为主不支持多剧集管理，以单条氛围感视频为主接入方式网页端为主，主体库与项目空间云端管理网页端 + 移动端协同，已接入 OpenClaw，可在飞书/钉钉/Discord 等聊天平台直接调用网页端 Dashboard 管理项目，支持多语言与主题切换生成规格支持多种画面比例（9:16/16:9/3:4/4:3），视频时长由分镜数量决定支持 16:9、9:16、1:1，单次最长生成 2 分钟完整成片未明确公开单次生成时长与比例细节目标用户短剧创作者、MCN 机构、知识博主等需要批量连续剧情的专业用户小白创作者、电商卖家、短视频博主、独立开发者，追求零门槛快速成片短视频创作者、社交媒体运营、音乐人、品牌营销团队，追求氛围感表达Seko的应用场景短剧漫剧创作：用一致性短剧模型与主体库锁定角色特征，批量生成跨分镜人物场景连贯的连续性剧情短片。音乐 MV 制作：上传音乐文件一键生成对口型视频，精准匹配情绪与节奏，打造电影质感的演唱或剧情 MV。知识科普宣传：通过知识分享模式快速制作科普讲解、政企推广、文旅宣传等全行业适配的口播与动画视频。广告商业短片：借助画布编辑与元素添加功能植入品牌 Logo 与商品图，生成带商业元素的营销与电商推广视频。二次精剪创作：导出独立分镜视频与独立音频文件，供专业剪辑软件进行深度合成与后期包装。©️版权声明：若无特殊声明，本站所有文章版权均归AI工具集原创和所有，未经许可，任何个人、媒体、网站、团体不得转载、抄袭或以其他方式复制发表本站内容，或在非我站所属的服务器上建立镜像。否则，我站将依法保留追究相关法律责任的权利。类似于Seko的工具亿话数字人对话智能体创作平台SkillHub腾讯云专为中国用户推出的 Skill 极速安装工具Suna全球首款通用型 AI Agent 开源项目AstronClaw科大讯飞推出的云端部署OpenClaw月亮树AI选品亚马逊AI选品智能体，亿级实时商品大数据遨虾1688推出的跨境电商生意Agent暂无评论再想想发表评论暂无评论…

JiuwenSwarm – 华为开源的 AI Agent 统一工作平台AI工具1天前发布AI小集03JiuwenSwarm是什么JiuwenSwarm是华为2012实验室、华为云、终端小艺等团队联合构建的开源AI Agent平台，也是首个面向鸿蒙PC的开源AI统一工作台。平台采用多智能体协同架构，支持办公、编程、娱乐等场景的一站式任务执行，创新推出 HITS人机协同新范式，让人与Agent共同组队完成复杂任务。平台全套开源，支持Windows、Mac、HarmonyOS、Ubuntu等多平台部署。JiuwenSwarm的主要功能集群模式办公：一句话组建多智能体团队，分头调研、梳理结构、补充内容、整合结果，20分钟生成200页PPT。智能编程开发：Code模式自主规划任务、完成ArkTS代码开发，支持前后端并行开发、分支合并与需求追加。人机协同游戏：支持五子棋、狼人杀等沉浸式互动游戏，智能体担任主持人或队友，实现人与Agent共同推理博弈。跨设备联动：与飞书、小艺等场景联动，支持手机远程唤醒鸿蒙PC上的智能体团队。多平台部署：支持Windows、Mac、HarmonyOS、Ubuntu等操作系统安装运行。微信关注回复“开源”，加入AI开源项目交流群如何使用JiuwenSwarm访问官网：访问JiuwenSwarm官网 https://openjiuwen.com/jiuwenswarm 进入下载页面，选择对应平台的安装包。安装部署：下载后按系统指引完成JiuwenSwarm客户端的安装与初始化配置。配置模型：在设置中接入所需的AI模型服务，确保智能体具备推理与执行能力。选择模式：根据任务类型切换办公、Code或娱乐模式，系统会自动匹配智能体团队。发起任务：输入需求描述，智能体团队自动拆解任务并协同推进，人可随时介入审核与调整。JiuwenSwarm的核心优势首个鸿蒙PC开源AI工作台：填补鸿蒙PC生态在AI统一工作台领域的空白，原生适配鸿蒙系统。HITS人机协同范式：创新提出”人在蜂群中”的协作模式，人不再是旁观者而是团队成员。多智能体并行加速：将串行任务拆分为多智能体同步推进，大幅提升办公与开发效率。全套开源可定制：代码完全开源，支持企业或个人基于AtomGit/GitHub进行二次开发与私有化部署。跨场景无缝衔接：覆盖办公、编程、娱乐三大场景，支持跨设备远程调度与多平台运行。JiuwenSwarm的项目地址项目官网：https://openjiuwen.com/jiuwenswarmGitHub仓库：https://github.com/openJiuwen-ai/jiuwenswarm/JiuwenSwarm的同类竞品对比对比维度JiuwenSwarmAutoGen开发背景华为2012实验室、华为云等联合构建微软研究院开源项目平台支持原生支持鸿蒙PC，覆盖Win/Mac/Ubuntu主要支持Windows/Linux，无鸿蒙原生版人机协同HITS范式，人可作为团队成员参与以Agent间协作为主，人机交互偏指令式场景覆盖办公、编程、娱乐一体化工作台偏重于编程与通用Agent编排部署方式提供鸿蒙PC专属安装包，开箱即用需通过Python pip安装，配置门槛较高开源协议开源，支持二次开发MIT开源，生态成熟JiuwenSwarm的应用场景紧急汇报材料生成：人在外出时通过手机飞书远程唤醒鸿蒙PC上的智能体团队，快速产出高质量PPT。鸿蒙应用快速开发：从需求描述到ArkTS代码编写、编译、安装运行，一站式完成鸿蒙原生应用开发。多人在线协作游戏：组织五子棋、狼人杀等互动游戏，智能体担任主持人或AI队友，增强社交娱乐体验。跨设备智能办公：与飞书、小艺深度联动，实现手机、PC、平板间的任务无缝衔接与持续推进。企业级多Agent调度：基于开源代码搭建私有化智能体协作平台，满足企业内部复杂业务流程自动化需求。# AI工具# AI项目和框架©版权声明本站文章版权归AI工具集所有，未经允许禁止任何形式的转载。上一篇KerWork - 桌面级AI执行助手，可在本地完成各种任务下一篇RingBot AI - AI智能体AaaS平台，零代码配置与上线相关文章海豚AI配音 – AI配音工具，支持真人音色和二次元音色AI小集6灵犀专业版 – 金山办公推出的独立 AI 办公 AgentAI小集2k1.5 – Kimi推出的多模态思考模型AI小集24Mem – AI笔记应用，自动识别笔记间的关联AI小集3畅问AI – 专家级AI助手，提供多角度、多领域的专业回答AI小集2瀚海智语 – 海洋垂直领域大模型，基于360 智脑和 DeepSeek 研发AI小集3暂无评论再想想发表评论暂无评论…热门工具豆包LibTV绘蛙AIAiPPT秘塔AI搜索TRAE编程堆友Agent美图设计室Sekoupdream秒哒办公小浣熊最新收录Lumina纳米Work袋马Sophclaw千问办公ReelFork最新文章秒悟Meoo – 1分钟生成前后端完整的网站，真0门槛开发！60分钟前Grok Voice Think Fast 2.0 – SpaceXAI 推出的语音模型1小时前AngelSpec – 腾讯混元团队开源的端到端推测解码训练框架1小时前Codex Security CLI – OpenAI开源的AI代码安全扫描CLI工具4小时前Lyria 3.5 – 谷歌 DeepMind 推出的 AI 音乐生成模型5小时前Qwen-Audio-Agent – 阿里开源的实时语音 Agent 框架21小时前NewMax – 本地 AI Agent 工作台，AI会真实操作浏览器23小时前RingBot AI – AI智能体AaaS平台，零代码配置与上线24小时前KerWork – 桌面级AI执行助手，可在本地完成各种任务1天前Instella-MoE – AMD 开源的混合专家语言模型系列1天前AgentLink – 心言集团推出的跨智能体工作流终端1天前ChatLens语镜 – AI 会话智能体，透镜读屏智能回复辅助2天前PerceptionBench – 月之暗面开源的视觉感知诊断基准2天前水杉输入法 – 免费开源的中文输入法，支持 AI 联想2天前Midjourney V8.2 – Midjourney推出的最新AI图像生成模型2天前