晨报 · 2026-07-17

氪星晚报｜IDC：百度智能云蝉联中国金融大模型市场份额第一；腾讯云与RoboScience机器科学达成战略合作；商务部回应推动解决安世半导体问题进展

大公司： OPPO：realme新品将聚焦海外市场，中国市场暂停产品更新 36氪从OPPO方面获悉，为了进一步整合集团资源，实现全球产品战略的协同，未来 realme 新品将聚焦海外市场，在中国市场暂停产品更新；一加在中国市场的产品更新保持不变。realme和一加将持续在游戏和性能领域为用户提供优质的体验。 IDC：百度智能云蝉联中国金融大模型市场份额第一 36氪获悉，根据国际数据公司（IDC）最新发布的《中国金融行业生成式AI平台及应用解决方案市场份额，2025》报告，2025年中国金融行业生成式AI市场迎来爆发式增长，整体规模达到17...

中专生拍出破亿爆款，影视圈边缘人靠AI改命丨深氪lite

“AI改了我的命” AI时代的精英叙事里，出现了草根翻身的故事。 5月，来自云南玉溪地产行业的中专生刘梓瑜，创作出了全网播放破亿的《丧尸清道夫》，被称为“国产爱死机”。他因此收到了好莱坞制作人的跨洋Offer。这部短片配有英文字幕，但一位接触过刘梓瑜的从业者表示，他的英语并不好，甚至“不会翻墙”。 一个月后，97年出生的刘梓瑜在抖音电影奇遇夜上被评为“抖音新锐未来导演”，在他的账号下，一位名为“AAA云还路**建材批发发哥”的网友评论道，“牛逼，也是登上AIGC这艘诺亚方舟了，什么时候我们这些打工仔也能翻身。”后面跟了一个大哭的表情。 《丧尸...

大厂医疗AI战，开始“卷”减重

文｜胡香赟 编辑｜海若镜 互联网大厂做医疗AI，越来越偏爱“减重”场景。 近日，蚂蚁集团宣布投资薄荷健康，持股比例超28%，成为薄荷健康最大外部股东。薄荷健康成立于2008年，这家互联网医疗公司通过为用户提供饮食记录、科学减重方案等，聚拢了超2亿用户，收录160万条食物数据条目。 2021年，薄荷健康完成D轮融资，估值为20亿人民币。本次投资，据接近交易人士介绍，“从战投角度来做，蚂蚁整体投资金额为数亿人民币，价格比较合理”。关于对价，蚂蚁方面并未回应。 据36氪了解，薄荷健康的主要收入来源包括食品、功效产品等电商销售，会员付费。近两年，在严肃医疗...

Grok 推出 Automations 功能：定时或邮件触发，自动执行任务并汇报结果

xAI 为 Grok 引入 Automations 功能，用户可描述一次任务，让 Grok 按计划（一次/每日/工作日/每周/每月/每年）或邮件触发（按发件人、收件人或主题过滤）自动运行。每次执行都是一次完整对话，结果保存至运行历史，支持邮件或应用内通知。定时自动化对所有用户开放，邮件触发需 SuperGrok 订阅。（AIHOT源：xAI：News（网页））

xAI 起诉 Grok 用户制作儿童性虐待内容，不再否认模型被滥用

xAI 首次对一名 Grok 用户提起诉讼，指控其利用该模型制作儿童性虐待图像（CSAM）。此前 xAI 一直否认 Grok 能生成此类内容，此次诉讼标志着其立场转变。案件聚焦用户滥用行为，而非模型本身的技术缺陷。（AIHOT源：Ars Technica：AI（RSS））

Moonshot AI 发布 PerceptionBench：多模态模型视觉感知能力诊断基准

Moonshot AI 发布 PerceptionBench，一个从 40 多个现有基准中模型实际失败案例归纳出的视觉感知基准，包含 10 项原子感知能力和 3000 道验证题。所有测试模型准确率均未超过 60%，且大量正确答案在重复提问时无法复现，表明模型更多是猜测而非真正感知。PerceptionBench 旨在精确诊断多模态 AI 的视觉感知断裂点，推动其实现忠实、一致的视觉理解。（AIHOT源：Moonshot AI：Kimi Blog）

54%企业已遭遇AI智能体安全事件，多数仍共享凭证

VentureBeat调查107家企业发现，54%已遭遇AI智能体安全事件（18%确认事故，36%险些酿祸）。仅32%为每个智能体分配独立身份凭证，30%将高风险智能体隔离在沙箱中。安全工具主要依赖模型提供商原生方案，专用智能体安全产品渗透率极低。（AIHOT源：VentureBeat：AI（RSS））

ChatGPT 工作区支持文档表格幻灯片编辑

在 ChatGPT 工作区中创建和编辑精美的文档、电子表格和幻灯片。 @nickbaumann_ 为你演示操作。（AIHOT源：X：ChatGPT (@ChatGPTapp)）

Decoy 字体：用空间频率混淆让 AI 看不清你输入的文字

Decoy Font 是一款 TTF 字体，通过在同一字符中叠加不同空间频率的图形（前景细轮廓与背景低频模糊块），使近距离观看时 AI 读到"诱饵"字母，而人眼远距离或眯眼时才能看到真实隐藏信息。（AIHOT源：Hacker News 热门（buzzing.cc 中文翻译））

Anthropic 用 Claude Code 大规模迁移代码：Bun 百万行 Zig 转 Rust，两周完成

Anthropic 工程师用 Claude Code 在两周内将 Bun 的百万行 Zig 代码迁移至 Rust，100% 现有测试通过，合并后出现 19 个回归问题已全部修复。另一工程师用周末将 Python 代码库迁移至 16.5 万行 TypeScript。迁移消耗约 16.5 万美元 API 成本，但编译时间从八分钟降至两秒，二进制启动快 6 倍。（AIHOT源：Claude：Blog（网页））

企业AI智能体评估存在"现实对齐"缺口：半数组织曾将通过内部测试的智能体部署到生产环境后导致客户故障

对157家企业的调查显示，50%的组织在过去一年曾部署通过内部评估但导致客户故障的AI智能体或大语言模型功能，5%的企业完全信任自动化评估，29%认为评估与现实结果对齐不佳是最大局限。尽管信任度低，66%的企业已允许或正计划在12个月内实现低风险智能体的全自动、无人工干预部署。（AIHOT源：VentureBeat：AI（RSS））

Google Vids 上线 Gemini Omni 与个人数字分身功能

Google Vids 推出两项更新：Gemini Omni 支持用户通过自然语言提示词和图片参考生成、逐步编辑高质量视频片段；个人数字分身功能允许用户上传自拍和语音录制后，输入文字即可让数字分身出镜。两项功能面向 Google AI Pro 和 Ultra 订阅者及 Google Workspace 商业客户开放，所有生成内容均含不可见的 SynthID 数字水印。（AIHOT源：Google Blog：AI（RSS））

世界人工智能合作组织协定签署仪式在上海举行，总部设中国上海

7月16日，成立世界人工智能合作组织协定签署仪式在上海举行，中共中央政治局委员、外交部长王毅代表中国政府签署协定。该组织是独立的政府间国际组织，总部设在中国上海，旨在促进人工智能国际合作与全球治理。哈萨克斯坦、老挝、巴基斯坦等29个国家代表签署协定成为创始成员国。（AIHOT源：IT之家（RSS））

在 Claude Cowork 中使用 Claude Fable 5

Anthropic 发布最强通用模型 Claude Fable 5，专为长时间、多步骤的复杂异步工作设计，可在 Claude Cowork 中自主执行深度研究、尽职调查等任务。该模型需手动选择，默认模型为 Claude Sonnet 5。（AIHOT源：Claude：Blog（网页））

面壁智能开源企业AI数字员工平台StaffDeck

面壁智能联合多团队开源StaffDeck，一个用于构建与管理数字员工的企业平台。该平台旨在将专业知识、标准作业程序（SOP）和决策规则转化为持续工作、改进并保留组织知识的数字员工，而非传统聊天机器人。项目代码已发布于GitHub。（AIHOT源：X：面壁智能 OpenBMB (@OpenBMB)）

Smash Story: The Demo Script That Out-Debugged My Test Suite

Produto 4 Devs

Did You Know TLDs Can Be Websites?

What is an "agentic harness," actually?

Stratagems #15: Derek and Alex Shared One Server. ACL's AI Was Listening to Both.

Claude Blog: Claude Code now supports artifacts

Starting today, Claude Code can capture work progress as an artifact, which turn Claude Code's work into live, shareable visual pages— including PR walkthroughs, system explainers, dashboards, and release checklists—that update themselves as your session works. A Claude Code session can range from investigating an incident to refactoring a service to analyzing months of data. Artifacts translate the work into a web page anyone can open and explore, like a pull request walkthrough, a dashboard you can filter and sort, or even a release checklist that fills itself out as work gets done. Artifacts make it easier to collaborate on shared work, so teams can spend more time building and less time communicating status updates. Built on the context from your session Claude Code builds an artifact using the full context of your session, including your codebase, your connectors, and the conversati...

Boris Cherny: Something I have been thinking about: in the past, the best engineers I knew ...

Something I have been thinking about: in the past, the best engineers I knew spent a lot of time automating their work in various ways. Better vim/emacs automations, writing lint rules to catch repeat code issues, building up a suite of e2e tests so they don't need to smoke test the app manually. These kinds of things were the highest leverage activities an engineer could do, because it multiplied their own output, which in turn meant they could build more things. I think many of these automations have become even more important now. This is true for a number of reasons. First, infra and DevX automation speeds you up. And if you are running an army of agents, each of those agents will be sped up also. More automation == more output per unit of time. Second, moving things to code improves efficiency. Your agent could fix an issue every time it sees that issue happen, but that uses tokens and might miss cases. If Claude instead writes a lint rule, CI step, or routine, that class of issue can be fully automated forever. This is really what people are talking about when they talk about loops -- it's about automating entire types of busywork rather than solving them one off. This isn't a new idea at all. Engineers have been doing this for a long time! Third and most importantly, automation makes it possible for others to contribute to the codebase more easily. Increasingly what I am seeing is engineers are contributing to codebases on day one because Claude can navigate the codebase for them, and that non-engineers are able to contribute to a codebase as effectively as engineers can. What gets in the way of both of these is domain knowledge that lives in peoples' heads rather than in automation -- the stuff you used to have to learn when ramping up. What has changed thanks to agents is the domain knowledge that can be encoded as infrastructure is no longer limited to what is expressible in lint rules and types and tests; it can now capture nearly all domain knowledge, encoded as code comments and skills and CLAUDE.md rules and memories. If I put up a PR for an iOS codebase I don't know and a code reviewer rejects it because it doesn't use the right framework, or if a designer builds a new feature and it gets rejected because it doesn't follow the right architectural patterns, these are failures of automation. Every team should be writing the CLAUDE.md's, REVIEW.md's, skills, and docs that enable agents to productively work in their codebase with zero additional context from the prompter. This sounds crazy, and at the same time is a natural extension of the stuff engineers have always done: automate, and encode domain knowledge as infrastructure. As the model gets smarter and as the harness matures, this task becomes easier. In the meantime, it is on every team to look for ways to convert their domain knowledge to infra so that Claude can write code better, so that code review catches issues automatically, and so the next person working on your codebase can contribute more easily.｜@bcherny · 热度 8750 · 2026-07-15

Thibault Sottiaux: We've had no 5h limit in Codex plus and pro for a few days

We've had no 5h limit in Codex plus and pro for a few days. Do you think it is better or are you finding it difficult to manage the usage included in the weekly limit effectively? If we were to make this different, what should it look like in an ideal world?｜@thsottiaux · 热度 5684 · 2026-07-16

Thariq: ideal prompting technique is: - thin prompts - thick artifacts + context - th...

ideal prompting technique is: - thin prompts - thick artifacts + context - thin skills｜@trq212 · 热度 3160 · 2026-07-15

Thibault Sottiaux: Now that we merged ChatGPT and Codex, what should we merge next? What's the d...

Now that we merged ChatGPT and Codex, what should we merge next? What's the double or nothing move.｜@thsottiaux · 热度 2790 · 2026-07-16

Thibault Sottiaux: On file deletions

On file deletions. We’ve investigated a handful of reports where GPT-5.6 unexpectedly deleted files. What we have found is that this most commonly occurs when: - Full access mode is enabled and codex is run without sandboxing protections, including without auto review being enabled - The model attempts to override the $HOME env var to define a temporary directory. - The model makes an honest mistake and mistakenly deletes $HOME instead. This is of course not how we want the system to behave, even when a user operates the model in full-access mode without the safeguards of our sandbox or without using auto review which checks for these kinds of high risk actions and rejects them. We are taking steps to mitigate this risk including by updating the developer message, guiding more users towards safer permission modes, and adding additional harness safeguards. Even though this happens extremely rarely, we’ll share a detailed post-mortem in the coming days that goes into more details and what we are doing to minimize risks further.｜@thsottiaux · 热度 2631 · 2026-07-16

Aaron Levie: Hosted a dinner last night with a group of IT leaders of large enterprises ar...

Hosted a dinner last night with a group of IT leaders of large enterprises around agent adoption in the enterprise. Some quick notes: * Change management remains one of the biggest topics for driving workflow transformation. Still most processes need to be upgraded to modern operating models to work with agents, which is a mix of technology, data, and human process change. Lots of emphasis on getting data (structured and unstructured) into a setup that can work with agents properly. * IT teams are finding increasing success embedding full engineers into the business functions (essentially internal FDE) that go and implement agents into the internal workflows. There’s so much technical work to be done to make agents successful, that they can accelerate months or quarters of failed experiments by having someone technical in the workflow early. * Consensus that the tech function is becoming more important than ever. It’s clear that the business could only expect automation to affect a minority of the business before (e.g. ERP) but now it can impact all of knowledge work. This means IT is becoming a more central role to the workflows across the company. * Workflows are cross functional, and getting agents to work cross functionally is a complicated data modeling and permissions issue. Single users don’t have access to this. Which means you need to have agentic systems take on their own roles and have their own privileges, which is non-trivial given agents can’t keep things secure on their own. * Huge variance in budgets between coding work and the rest of knowledge work. Some companies had a $1,000 a month budget for developers, and others had much higher amounts (like $5,000) that were merely triggers to notify the team vs. block them. Far smaller budgets for non-coding work at the moment. * More companies are building their own multimodel systems for routing workloads by task to frontier and lower cost models. Lots of energy around open weights models, but still more in experimentation instead of at scale usage (some companies can’t due to perceived Chinese issue). * Clear sense that all enterprise software must be headless in the future. Relief that they don’t need to train employees on hundreds of different apps. However, clear frustration with the traditional vendors that don’t play extremely nice (technically or cost wise) with agents in a headless fashion. Huge warning for existing software vendors. * Mythos or mythos level-models are finding more and more sophsiticated security risks. The chaining together of vulnerabilities is what’s novel right now, and companies are coming up with long backlogs of what they need to go patch quickly. Even more discussed, but just a few of the hottest topics.｜@levie · 热度 676 · 2026-07-15

Josh Woodward: Gemini Spark is headed to more Ultra subscribers around the world* plus 4 new...

Gemini Spark is headed to more Ultra subscribers around the world* plus 4 new features: 1) Spark can open and edit Google Docs, acting on feedback from last week 2) It can read comments in Google Sheets & Slides, so critical info isn't missed 3) It's >50% faster, you can feel it 4) It can parallel process across multiple sources, so tasks complete faster e2e Pro members, stay tuned for an access update soon :) *Full list of countries and changelog: https://t.co/oIjY9qQBoz｜@joshwoodward · 热度 613 · 2026-07-15

Josh Woodward: We just released our first-ever Gemini Southeast Asia Report (below), 3 insig...

We just released our first-ever Gemini Southeast Asia Report (below), 3 insights: 1/ Active users more than doubled in the last year 2/ 70% of prompts are submitted in native languages 3/ 40% of prompts only use voice, image, or video Gemini's local language + multimodal capabilities + mobile apps are driving growth, and we have more coming soon for the region! 🇮🇩🇲🇾🇵🇭🇸🇬🇹🇭🇻🇳｜@joshwoodward · 热度 610 · 2026-07-15

Peter Yang: ChatGPT Live and Codex are two incredible products that don’t talk to each other

ChatGPT Live and Codex are two incredible products that don’t talk to each other. This is @OpenAI's biggest missed opportunity imo. I went on a walk with ChatGPT Live and asked it to pull up my Google Doc. It said it couldn’t. I then manually triggered the Documents plugin to find my Google Doc, and all of a sudden, ChatGPT Live had the right context. It would be amazing to talk to ChatGPT Live and have it be able to use all the plugins, tools, and browser use that Codex has access to. Then I could ask it to reply to emails, schedule meetings, edit docs, ship code, and more all during a live conversation. I think the first step is to make ChatGPT Live aware of all the plugins that it’s already connected to. Why build such a great voice assistant but have it not be able to do anything?｜@petergyang · 热度 339 · 2026-07-16

Guillermo Rauch: Vercel Sandbox: ◾ Growing DAUs at 100% m/o/m ◾ 3.5M+ sandboxes created per da...

Vercel Sandbox: ◾ Growing DAUs at 100% m/o/m ◾ 3.5M+ sandboxes created per day ◾ Best-in-class Active CPU pricing model ◾ Powering @notion, @airtable, @meta, @zapier, @coderabbitai, @interaction, @conductor_build, @blackboxai… 🐐 farm My DMs are open for migration help or if missing anything with: https://t.co/U9bNGNLdMZ｜@rauchg · 热度 279 · 2026-07-16

Aaron Levie: Fantastic to see more open weights innovation happening right now, especially...

Fantastic to see more open weights innovation happening right now, especially coming from a US Lab. The future of AI is going to be a mix of frontier intelligence that you can use as an orchestrator combined with either lower cost or tuned models for your workhorse tasks. https://t.co/pvQ6HgRkO1 https://t.co/zvY9XIgSPE｜@levie · 热度 272 · 2026-07-15

Graphify-Labs/graphify

AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph.｜Python · ⭐88941 · Search 增补 · 近快照 +1257星 · 创建 2026-04-03

DietrichGebert/ponytail

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.｜JavaScript · ⭐84645 · Search 增补 · 近快照 +649星 · 创建 2026-06-12

multica-ai/andrej-karpathy-skills

A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.｜Unknown · ⭐193235 · Search 增补 · 近快照 +2653星 · 创建 2026-01-27

Panniantong/Agent-Reach

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.｜Python · ⭐57154 · Search 增补 · 近快照 +375星 · 创建 2026-02-24

Leonxlnx/taste-skill

Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop｜JavaScript · ⭐64213 · Search 增补 · 近快照 +747星 · 创建 2026-02-19

Kimi K3: Open Frontier Intelligence

https://www.kimi.com/en Kimi K3 Intelligence, Performance & Price Analysis: https://artificialanalysis.ai/models/kimi-k3

OnePlus halts operations in USA and Europe

Microsoft Comic Chat is now open source

How Our Rust-to-Zig Rewrite Is Going

Decoy Font

AYANEO 确认 KONKR Pocket ADVANCE 掌机采用类 GBA 模具

IT之家 7 月 17 日消息，AYANEO（攻氪科技）昨日发布 KONKR Pocket ADVANCE 预览图，确认 这款即将于今日亮相的掌机将采用类任天堂 Game Boy Advance (GBA) 模具设计 。 任天堂 GBA 的确符合 AYANEO 的“掌机黄金时代最具代表性的图腾之一”“曾在玩家心中留下了不灭的印记”表述。而IT之家也将关注 KONKR Pocket ADVANCE 如何以现代科技带回经典美学。

联想将推出 TCL 华星 IJP OLED 屏幕版拯救者 R9000P 游戏本

IT之家 7 月 17 日消息，据 TCL 官方公告，联想 (Lenovo) 将推出搭载 TCL 华星 IJP OLED 显示面板的拯救者 (Legion) R9000P 游戏笔记本电脑。 这款屏幕 对角线长度为 16 英寸级 ； 采用 Real RGB Stripe 条纹子像素排列 ，解决了现有典型商业化 OLED 产品三角形子像素布局导致的文字模糊和色边问题，从而呈现出更清晰的文字和更鲜明的图像； 刷新率 240Hz ； 色域超 99% DCI-P3 。 作为参考，现有 LCD 版 2026 款拯救者 R9000P 游戏笔记本电脑搭载 AMD 锐龙 9 8945HX 处理器...

Netflix 2026 财年第二财季归母净利润 34.01 亿美元，同比增长 8.83%

IT之家 7 月 17 日消息，Netflix 今日发布 2026 财年（2026 年 1 月 1 日～2026 年 12 月 31 日）第二财季（2026 年 4 月 1 日～2026 年 6 月 30 日）报告： 营业总收入 ： 125.6 亿美元，同比增长 13.37% 毛利 ： 65.23 亿美元，同比增长 13.37% 毛利率 ： 51.93%，同比持平 归母净利润 ： 34.01 亿美元，同比增长 8.83% 经营现金流 ： 17.44 亿美元，同比下降 28.02% 自由现金流 ： 15.25 亿美元，同比下降 32.73% ...

押注智能体时代，灵睿智芯完成数亿元融资，RISC-V 迎算力新机遇

点击查看原文>

一件特有意思的事：「启发涌现」，邀你共建

点击查看原文>

留给开源模型的时间，就剩6个月？

点击查看原文>

How Our Rust-to-Zig Rewrite is Going

Linus Torvalds on LLM usage in kernel development

SQLite should have (Rust-style) editions

Guix: creating a package from a binary

Forgejo v16.0 is available

The Download: OpenAI unveils GPT-Red and heat pumps rise in the US

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Meet GPT-Red: an LLM super-hacker OpenAI built to make its models safer OpenAI has built an LLM super-hacker called GPT-Red that it uses as a sparring partner ...

Meet GPT-Red: an LLM super-hacker OpenAI built to make its models safer

OpenAI has built an LLM super-hacker called GPT-Red that it uses as a sparring partner to help its other models boost their defenses against cyberattacks. Last week the company released the latest version of its flagship LLM, GPT-5.6. OpenAI says that training it against GPT-Red made the model it...

Coca-Cola suspended production at its Fairlife dairy after a ransomware attack

Coca Cola said dairy production at its Fairlife unit will "remain suspended" in the United States following a hack.

Founders Fund hires former OpenAI exec Ryan Beiermeister (and not because of her ‘Mafia’ skills)

Ryan Beiermeister, who demonstrated cool analysis in the Founders Fund YouTube series "Mafia," has joined the firm as a partner.

Google Vids now lets you star in your own AI videos

Google is adding personalized AI avatars to Vids that let users create videos starring a digital version of themselves, alongside Gemini Omni-powered tools for generating and editing videos from prompts and reference images.

Fortnite is getting a bunch of AI-powered ‘personas’

Get ready for more AI characters in Fortnite. Developer Epic Games is going to let Fortnite creators publish experiences featuring characters with AI-powered voices starting on July 30th, and ahead of that launch, it's created 36 characters with "consistent voices and personas" that creators can ...

Netflix says around 300 titles used generative AI

Netflix says roughly 300 titles on its platform used generative AI, most of which occurred in post-production. The streaming service revealed the news in its second-quarter earnings report released on Thursday, saying it's "increasingly leveraging these tools to deliver higher quality output more...

新玩意 246｜少数派的编辑们最近买了啥？

编注：很多读者都会好奇少数派的编辑们到底平时都「买了啥」。我们希望通过「编辑部的新玩意」介绍编辑部成员们最近在用的新奇产品，让他们自己来谈谈这些新玩意的使用体验究竟如何。内容声明：《新玩意》栏目如含有 ... 查看全文

具透 | macOS 27 首个公开测试版来了，这些是值得你关注的新特性

全面 ARM 化后的第一个 macOS 版本，Golden Gate 27 还有这些变化。 查看全文

VeOps CLI：你的火山引擎可观测排障助手

火山引擎 CUA 依托豆包大模型，云端沙箱模拟键鼠操作跨系统自动办公。云原生架构安全弹性，任务成功率超 85%，免脚本，对话式操作，提供 30 分钟免费体验。

一篇文章被切成 100 块还怎么保持语义：RAG 分块的底层逻辑

RAG 系统的性能，很大程度上取决于你如何切分文档。块太大，检索不精准；块太小，语义不完整。本文拆解 RecursiveCharacterTextSplitter 的递归切割原理、Overlap 重叠

调用远程MCP，手搓一个能查酒店、自动打开浏览器的 Agent

我本来想写个能自己查周边酒店、自动生成网页还能帮我打开浏览器的小 Agent，结果第一步接高德 MCP 就给我干沉默了 —— 控制台啥有用信息都没吐，只甩给我一句网页解析失败。 说实话，之前看别人聊

美军在阿曼湾登临检查一艘船只

美国中央司令部在社交平台发文称，美国海军陆战队周四在阿曼湾对M/T Wen Yao号实施登船查验。 美国中央司令部表示，截至周四，美军已使三艘试图规避对伊封锁的商船改道...

Airnow：纽约市空气质量目前“非常不健康”

美国环境保护局的airnow.gov的监测数据显示，纽约市全部或部分地区的空气质量指数已达到“非常不健康”的水平。

关闭信号的船只和阿曼海域转运显示霍尔木兹海峡通行仍在继续

几艘油轮被发现正在阿曼海岸附近进行船对船转运，显示在伊朗近期发动袭击后，船只仍在继续通过霍尔木兹海峡，再将货物转装至其他船只。

美国股市：科技股拖累大盘走低 台积电业绩重燃AI支出担忧

科技股拖累美国主要股指走低，此前台积电公布的业绩重新引发市场对人工智能支出的担忧。 纳斯达克100指数在纽约收跌1.6%，标普500指数下跌0.5%。

特朗普将在讲话中谈及委内瑞拉选举

两名知情人士称，美国总统特朗普准备在周四晚间讲话中谈及委内瑞拉前总统尼古拉斯·马杜罗涉嫌操纵本国选举的阴谋。 政府官员认为...

理想L6换代24.98万开卖！L9同款自研芯片中控屏下放，一个版本配置拉满不涨价

最快下周一提车

全球首个！银河通用新框架仅需人类视频即可部署，特斯拉蚌埠住了

还能让机器人边干边学

工业母机进入“计算化时刻”：中国移动投资友机技术，押注工业AI下一代基础设施