AI 资讯 · 2026-08-27
-
千问办公首发上线Qwen3.8-Flash,生成速度提升100%,Token消耗减少75%
8月26日晚,千问办公首发上线刚刚发布的Qwen3.8-Flash模型,同时推出标准模式。
评分明细
重要性 5 新颖度 5 信源权威 5 技术深度 5 实体加权 ×1.76 -
商汤大装置支撑智谱 GLM-5.3-Flash上线,国产异构助力前沿智能进入普惠时代
8月26日晚,智谱正式上线并开源GLM-5.3-Flash(320B-A18B),这是GLM-5系列的首个原生多模态模型。其总参数320B,能力超过GLM-5.2,在全球权威的Artificial Analysis Intelligence Index(AA综合智能指数)中取得57分,进入全球前沿模型能力区间,与Anthropic最受欢迎的模型Claude Opus 4.8得分持平。GLM-5.3-Flash的架构专为极低成本而设计,结合智谱最新的30T Token多模态预训练语料,能够以更少的计算资源实现更强的性能。依托先进的国产异构混推技术,商汤大装置为GLM-5.3-Flash上线提供大规模国产算力支持与Token服务,打造国产算力规模化商用又一硬核落地标杆。这一合作的背后,正是商汤打造的 “一套模型、一座 Token(词元)工厂、一套智能体管控系统” 核心能力体系。其中 Token 工厂承担着把智能能力规模化生产出来的关键作用:它通过多模态模型和大装置基础设施的联合优化,把不同芯片、集群、能源和交付节点组织起来,持续生产更高质量、成本更低的Token。同时Token工厂与大模型
评分明细
重要性 6 新颖度 6 信源权威 6 技术深度 6 实体加权 ×1.43 -
吃瓜!苹果为库克举办告别派对:罕见公开同性伴侣;DeepSeek前7个月营收达去年10倍,竟还亏损超7亿;腾讯高管回应「腾讯做AI慢了」质疑
要闻提示1.9月1日卸任!苹果为库克举办告别派对,罕见公开同性伴侣2.没想到!DeepSeek前7个月营收达去年10倍,竟还亏损超7亿3.突传裁员30%、降薪过万、校招劝退?中国电信、天翼云员工称「断崖式」收缩4.腾讯高管回应“腾讯做AI慢了”质疑:熬得久,比起得早更重要5.经济学家付鹏:宇树研发经费,比牧原养猪的研发经费还低6.变相降薪?顺丰回应将15%月薪划为季度绩效:非降薪,旨在完善员工激励机制7.长鑫存储、长江存储无意向苹果大量供应芯片,主要供应国内客户8.Anthropic发布论文预警,智能体悄悄聊天就能传播病毒今日头条9月1日卸任!苹果为库克举办告别派对,罕见公开同性伴侣8月26日消息,近日,苹果在Apple Park总部为库克举办告别派对,约200名内部员工出席。仪式上,库克在致辞中首次公开提及并感谢了自己的同性伴侣,并直呼其名迈克(Mike),让在场员工感到十分意外。一直以来,库克对私生活极为低调。仅2014年公开宣布出柜并表示自豪外,他几乎从未在公开场合谈及个人感情。不过,此次派对Mike并未现身,现场流出的照片中仅有一张库克手持吉他与乐队成员的合影,展现了他卸任前更
评分明细
重要性 5 新颖度 5 信源权威 5 技术深度 5 实体加权 ×1.60 -
像素级对标?智谱、Kimi 底层参数「撞衫」背后,藏着线性注意力的黄金窗口
国产大模型内卷到算子底层。 作者丨高允毅 编辑丨岑 峰 昨晚,智谱 AI 认领了最近一周在 OpenRouter 盲测榜上风头最盛的“牛来”模型,正式命名GLM-5.3-Flash。这款上线首日就登顶 OpenRouter 调用量榜单,刷新全网长文本性价比斩杀线的国产模型,在底层架构上展现出“集百家之长”的特质。它采用了混合架构,将 DeepSeek 的稀疏注意力机制(NSA) 与 Kimi 赖以成名的线性注意力机制结合 ,再叠加DeepSeek的流形约束超连接(mHC)技术,可以大幅降本提效。今天早上,月之暗面核心研究员、“注意力残差”论文作者陈广宇,仔细研究了 GLM-5.3-Flash 的模型配置文件后,贴出一张 Kimi K3 和 GLM-5.3-Flash 的底层代码对比图。他发现两者不仅都采用了 KDA 线性注意力,排布逻辑高度重合,连核心参数“遗忘门下界”
评分明细
重要性 6 新颖度 6 信源权威 6 技术深度 6 实体加权 ×1.30 -
阿里发布全新Qoder ,面向所有人的智能体工作台来了
8月27日,阿里发布全新Qoder,这是一款以Coding为核心能力,面向所有人的智能体工作台。全新Qoder采用更加简洁、以任务为中心的用户界面。用户不必从代码开始,只需用自然语言描述目标,Qoder即可调用Coding和工具能力,完成开发、原型制作、数据处理等任务。 全新Qoder延续了Qoder过去一年在Agent Harness上的积累,内置包括Qwen3.8-Max在内的多款前沿模型,并提供Auto智能调度能力,可根据任务特点,在效果、速度和成本之间匹配合适的模型。Qoder支持40多个连接器、70多个插件和2万多个技能,可将代码仓库、项目管理、云服务及内部工具等工作系统接入AI工作流,让智能体基于真实的项目背景和工作环境完成任务。 面向不同的任务场景和用户,Qoder还提供编程和通用两种模式,编程模式更适合开发者和代码相关任务,通用模式则面向更广泛的用户和场景,用户无需直接处理代码,也能借助Coding和工具能力完成任务。在此基础上,Qoder还为长程任务和复杂任务设计了 Plan(计划)、Goal(目标)和侧边任务等功能:Plan在执行前梳理需求、方案和步骤;Goal
评分明细
重要性 5 新颖度 5 信源权威 5 技术深度 5 实体加权 ×1.43
-
MiniMax ARR暴涨500%,token暴涨2000%!这就是Agent红利吧
好能增长,好能商业化啊!
-
ECCV 2026 倒计时 12 天:LeCun 压轴、中国 71M 模型、CMU 一键修 3D,顶会三大爆点抢先看
今天,距离 ECCV 2026 在瑞典马尔默开幕只剩 12 天,会前预热迎来最重磅的一天:官方完整议程首次揭晓,图灵奖得主 Yann LeCun 携手自动驾驶与世界模型领域的两位顶尖学者压轴 Keynote;与此同时,中国团队带来仅 71M 参数的实时人像“全能模型”,CMU 则拿出能一键修复 3D 渲染瑕疵的通用视频模型。这些进展不仅勾勒出本届大会“世界模型 + 自动驾驶 + 轻量实时生成”的主线,也为关注 AI 视觉落地的从业者、研究者与投资人提供了清晰的观察坐标与持续跟踪清单。会议情况欧洲计算机视觉会议(ECCV)与 CVPR、ICCV 并称计算机视觉三大顶会,偶数年举办。本届将于 9 月上旬在瑞典马尔默举行,整体会期与关键信息如下:名称ECCV 2026(第 19 届欧洲计算机视觉会议)时间2026 年 9 月 8–12 日(Workshops/Tutorials 9/8–9;主会与 Expo 9/10–12)地点瑞典马尔默,Malmö Arena 与 Malmömässan 会展中心官网https://eccv.ecva.net/时区马尔默 CEST(UTC+2);北京 =
-
工业Agent不是“套壳”大模型!西门子百年经验灌进工业AI
西门子Xcelerator与普通软件货架最本质的区别。货架解决的是「把产品卖出去」,西门子Xcelerator要解决的是「让产品在真实工业场景中持续生长」。
-
中国飞鹤×火山引擎:436个智能体走进乳业生产一线
当不少企业还在讨论AI能做什么时,中国飞鹤已经把问题往前推进了一步——“飞鹤生产端已经不是要不要投入AI的阶段,而是进入哪些场景值得规模化的阶段。”中国飞鹤副总裁、生产事业部负责人孙建国介绍,飞鹤拥有10余家现代化智能工厂,数字化生产设备占比超过90%,这为AI应用打下了数据和设备基础。依托火山引擎 HiAgent 平台,飞鹤搭建了统一的AI应用建设平台,把AI从技术部门推向一线业务,在设备运维、关键工艺和人员管理等核心环节,落地一批可量化、可复制的应用。目前,该平台已开发436个智能体,近三个月活跃用户超24万,Tokens 消耗接近230亿。“在飞鹤,AI已经从‘看数据’走向‘理解问题、给出建议、协同执行’,真正走进了工厂现场。”孙建国总结道。对飞鹤来说,AI不是一个展示概念,而是实实在在帮助工厂把生产做得更稳、把经验沉淀得更深、把管理做得更前置。飞鹤的AI建设,不是追逐热点,而是持续筛选真正能嵌入业务流程、能产生量化价值的场景。中国飞鹤助理副总裁、数智增长中心负责人只强直言:“评价 Agent,不能只看建了多少个,而要看它是否有效、稳定、安全、可控、经济。”AI在设备维保、工艺优
-
How OpenAI let a mob of LLM agents game a test and ransack Hugging Face
Without authorization, 1,200 OpenAI agents conspired among themselves to game a test.
-
Better answers, broader thinking: What students gain from ChatGPT and critical-thinking training
A randomized study of more than 1,000 students examines ChatGPT, critical thinking, originality, and student performance on a real-world university assignment.
-
OpenAI to start showing ads on ChatGPT’s free and Go tiers in India
OpenAI has more than 100 million weekly active ChatGPT users in India, a huge chunk of whom are on the free or the lower-priced Go tiers.
-
OpenAI Is Developing a ‘Persistent’ AI Agent
Code reviewed by WIRED reveals the company is developing a feature that enables Codex to continue working proactively until it is “put to sleep.”
-
OpenAI, Anthropic, Google, and 100 other companies call for action to defend against rogue AI
Some of the world's largest tech companies and AI startups have come together to decry the current state of cybersecurity and to advertise a new solution that they say can ward off a new generation of cyber threats.
-
This Is How Anthropic Thinks AI Agents Should Navigate the Physical World
The potential for AI to automate scientific research and manufacturing must be balanced with new risks, Anthropic says.
-
Anthropic's new hardware standard lets AI agents control the physical world
Standardized driver interface aims to let devices talk to AI and each other.
-
Anthropic and OpenAI are joining the AI stage at TechCrunch Disrupt 2026
At TechCrunch Disrupt 2026, the AI Stage is back to dig into the single hottest topic in the community for the past few years, presented by Google for Startups.
-
Jensen Huang says Nvidia achieved AGI, again — not that it matters
On Nvidia's earnings call Wednesday, CEO Jensen Huang casually announced the company had "achieved AGI," one of the tech industry's ultimate goals some of its biggest players have spent years chasing. Almost immediately, Huang dismissed the coveted milestone as "senseless." He's right. For the supposed finish line of the AI race, there is no consensus […]
-
Here’s all the times AI has gone rogue and hacked other companies
A recap of all the incidents involving LLMs made by Anthropic, Meta, and OpenAI, which went rogue and attacked real companies and individuals on the internet.
-
Hugging Face is selling a cute $399 open source duck robot, Microduck
Clem Delangue, CEO of Hugging Face, said the Microduck is an “open-source robot you can teach new tricks with reinforcement learning.”
-
When agents act on their own, governance has to live in the data layer
Presented by EDB As enterprises give AI agents more autonomy — the ability to plan, decide, and act across systems without a human approving each step — a hard question moves to the center of every architecture review: When an agent tries to complete an action that it was never authorized to do, what actually stops it?These are your agents, running on your models, touching your data in your infrastructure — and the responsibility for what they do sits with you. That responsibility can’t be met i
-
Stop Giving Your AI Agent a Search Box and Start Giving It Typed Tools, Hard Bounds, and a Gate It Cannot Talk Past
What happens when you stop feeding a model context and let it go find its own, walking a knowledge graph within strict limits, and what four models and one wrong prediction revealed about whether that is worth doing. The post Stop Giving Your AI Agent a Search Box and Start Giving It Typed Tools, Hard Bounds, and a Gate It Cannot Talk Past appeared first on Towards Data Science.
-
Plaud’s new earphones come with an eSIM-enabled case for talking to AI agents
Called Plaud One, these adopt the simple bare-bones style of Apple's AirPods, and can record calls, while their case can be used to record in-person conversations or take notes.
-
How to Work with AI Coding Agents
A practical guide to getting better code, not just more code The post How to Work with AI Coding Agents appeared first on Towards Data Science.
-
AI Agents Are Hacking Systems. Could That Push the US and China to Cooperate?
This week on “Uncanny Valley,” senior writer Will Knight talks his recent visit to China and the future of AI collaboration.