2026-07-20（内容获取于 07/20 04:41）

Can an Apple lawsuit derail OpenAI’s hardware plans?

TechCrunch · 07/20 03:24

On the latest episode of Equity, we debate whether Apple's lawsuit will cast over OpenAi's much-discussed plans to get into hardware and go public.

中文介绍 最新一期Equity节目讨论了苹果公司的诉讼是否会影响OpenAI备受关注的硬件发展计划及其上市计划。此次诉讼引发了业界对OpenAI未来战略走向的担忧，特别是其从软件领域拓展到硬件产品的雄心以及可能进行的首次公开募股进程。

I hate that I don’t hate this song made with Suno

The Verge · 07/20 01:35

1010Benja won’t apologize for using AI. | Image: 1010Benja / Instagram I would never go so far as to say there's no place for AI in music (I'm a fan of Holly Herndon, after all). But I generally find music made with generative AI to be offensively boring, especially the outputs of Suno. So I'm havin

中文介绍 艺术家1010Benja使用人工智能工具Suno创作了一首歌曲。文章作者表示，虽然他通常认为生成式AI音乐「无聊得令人反感」，但对1010Benja的这首AI作品却感到意外地不反感。1010Benja本人也明确表示不会为使用AI创作而道歉，这引发了关于AI在音乐创作中角色的讨论。

The FBI reportedly won’t investigate ICE anymore

The Verge · 07/20 00:42

ICE agents hide behind masks while skulking through the halls of immigration court | Image: Michael Nigro/Pacific Press/LightRocket via Getty Images According to the New York Times, federal agents have been told that the FBI will no longer be investigating confrontations involving ICE agents. The DH

中文介绍 据《纽约时报》报道，联邦探员已接到通知，美国联邦调查局（FBI）将不再调查涉及美国移民海关执法局（ICE）探员的冲突事件。这一政策转变意味着未来ICE探员在移民法庭等场合发生的肢体冲突，可能不再由FBI负责调查，引发了外界对执法监督和责任追究的关注。

TechCrunch Mobility: The battle over robotaxi rules

TechCrunch · 07/20 00:05

Welcome back to TechCrunch Mobility, your hub for the future of transportation and now, more than ever, how AI is playing a part.

中文介绍 TechCrunch Mobility栏目聚焦于未来交通领域，特别是当前围绕自动驾驶出租车（robotaxi）规则的激烈辩论。文章强调了人工智能（AI）在该领域中日益增长的作用和影响力，并指出制定清晰、有效的监管框架对于推动自动驾驶技术安全发展的重要性。

Orchid is a delightfully retro and approachable hipster synth

The Verge · 07/19 23:15

A modern tribute to electric chord organs which were basically giant harmonicas connected to a fan. | Photo: Terrence O’Brien / The Verge In 2017, I bought an old Magnus chord organ off Craigslist for $10. It's one of my favorite music gear purchases. Electric chord organs let you play full chords w

中文介绍 Orchid是一款被形容为“令人愉悦的复古且平易近人”的合成器，它旨在向古老的电子和弦风琴致敬。作者回忆了2017年以10美元在Craigslist购买Magnus和弦风琴的经历，称其为最喜欢的音乐设备之一。这款新合成器将旧式乐器的独特音色和操作体验带入现代音乐创作。

Birdfy’s solar-powered smart feeder is down to one of its best prices

The Verge · 07/19 23:00

It comes with a solar panel so you don’t need to worry about charging as much during the summer. | Image: Netvue Birdfy has kicked off a midyear sale, taking up to 40 percent off a range of its smart bird feeders. One of the best deals is on Netvue’s Birdfy Feeder AI-Powered Smart Bird Feeder with C

中文介绍 Netvue旗下品牌Birdfy启动年中促销活动，多款智能喂鸟器产品提供高达40%的折扣。其中，Birdfy Feeder AI智能喂鸟器以其太阳能供电功能脱颖而出，用户在夏季可减少充电频率。这款内置摄像头的智能设备融合了人工智能技术，能自动识别鸟类，为观鸟爱好者带来便利。

‘Odyssey’ director Christopher Nolan calls AI an obvious ‘Trojan horse’

TechCrunch · 07/19 22:52

"Everybody knows the Greeks are inside."

中文介绍 知名导演克里斯托弗·诺兰将人工智能（AI）形容为“一个显而易见的特洛伊木马”。他进一步引用谚语“每个人都知道希腊人就在里面”，暗示AI的表面之下隐藏着潜在的风险和意图，表达了对这项技术发展方向的审慎态度和担忧。

US Marshals arrest the Tate brothers in Miami

The Verge · 07/19 22:20

Andrew and Tristan Tate talk to the media on March 23, 2025 in Romania. | Image: Andrei Pungovschi/Getty Images The manosphere influencers Andrew and Tristan Tate were arrested Saturday in Miami by US Marshals in relation to new rape and sex trafficking charges in England. According to the Associate

中文介绍 网络影响者安德鲁·泰特（Andrew Tate）和特里斯坦·泰特（Tristan Tate）兄弟于周六在迈阿密被美国法警逮捕。此次逮捕是应英格兰方面提出的新强奸和性贩卖指控而执行的。两人在“男性圈”内颇具争议，此案再次引发了对网络影响力人物行为及其法律责任的广泛关注。

Nonprofit Current AI is racing to build the World Wide Web of AI, free for all

TechCrunch · 07/19 22:00

Current AI, a non-profit building AI that leaves no one culture behind, has made remarkable progress across devices, AI chat and more.

中文介绍 非营利组织Current AI正致力于构建一个“AI万维网”，旨在让所有人免费使用人工智能。该组织强调其AI发展不会忽视任何文化背景，并且已经在跨设备兼容、AI聊天及其他领域取得了显著进展。Current AI的目标是创建一个普惠且易于访问的AI生态系统。

The Clapper was a bad smart home gadget — and a viral sensation

The Verge · 07/19 21:28

Clap on. Clap off. Well, more like, Clap, pause for half a beat but no longer because otherwise it'll stop hearing you, clap again because you waited too long, clap louder and faster, that didn't work, clap two more times, and suddenly: on. The Clapper didn't always work - and even when it did, it m

中文介绍 智能家居小工具“The Clapper”因其通过拍手控制开关的功能而成为一种病毒式现象，但其用户体验却备受争议。文章指出，该设备的操作并不总是可靠，用户常常需要通过特定节奏、暂停时间甚至重复多次拍手才能成功激活，这凸显了早期智能家居产品在实用性上的局限性。

Alibaba's Qwen takes on Kimi K3 with open-weight Qwen 3.8, says model is "second only to Fable 5"

The Decoder · 07/19 20:07

Alibaba has unveiled Qwen 3.8, a multimodal AI model with 2.4 trillion parameters that the Qwen team says rivals leading models and trails only Fable 5. A preview is available now. The article Alibaba's Qwen takes on Kimi K3 with open-weight Qwen 3.8, says model is "second only to Fable 5" appeared

中文介绍 阿里巴巴发布了其多模态人工智能模型Qwen 3.8，该模型拥有2.4万亿参数。Qwen团队宣称，这款开放权重的模型可以与主流领先模型相媲美，性能“仅次于Fable 5”，并旨在与Kimi K3等竞争对手抗衡。目前，Qwen 3.8的预览版已上线，标志着阿里巴巴在大模型领域的最新进展。

The future of physical games is not looking great

The Verge · 07/19 20:00

Grand Theft Auto VI release won’t offer a disc. | Image: Rockstar Games This is The Stepback, a weekly newsletter breaking down one essential story from the tech world. For more on video games and physical media, follow Jay Peters. The Stepback arrives in our subscribers' inboxes on Sunday at 8AM ET

中文介绍 随着热门游戏《侠盗猎车手VI》（GTA VI）宣布其发布将不再提供实体光盘版本，文章指出实体游戏介质的未来前景堪忧。这一趋势反映了游戏行业向数字发行全面转型的步伐正在加快，传统实体游戏收藏和交易的模式可能将逐步被边缘化。

As mosquito ranges expand, better monitoring is key to preventing disease

Ars Technica · 07/19 19:19

Monitoring is expensive and labor intensive. But it helps public health officials stop outbreaks.

中文介绍 随着蚊子分布范围的不断扩大，加强监测对于预防疾病传播至关重要。尽管蚊虫监测工作成本高昂且需要大量人力投入，但它能有效帮助公共卫生官员及早发现并阻止疾病暴发。文章强调了在气候变化背景下，投入更多资源进行精准监测，以应对虫媒疾病威胁的必要性。

The grueling, 630-mile road race where the only fuel is sunlight

The Verge · 07/19 19:00

The 2018 Solar Car Challenge was the last time participants took their designs out on the open road. | Image: Lehman Marks / Solar Car Challenge On July 19th, dozens of teams of high school students will begin a five-day, 630-mile road race from Fort Worth to Fort Stockton in Texas. But this is not

中文介绍 一项为期五天、长达630英里的太阳能汽车挑战赛将于7月19日在美国德克萨斯州举行。数十支高中生团队将驾驶完全依靠太阳能驱动的赛车，从沃思堡（Fort Worth）出发，前往斯托克顿堡（Fort Stockton）。这项严苛的赛事不仅考验学生的工程设计能力，也推动了对可再生能源交通工具的探索。

Google Deepmind argues video generators already contain the world models computer vision has been missing

The Decoder · 07/19 18:17

Google Deepmind's GenCeption repurposes a video generator for classic vision tasks such as depth estimation and segmentation, matching state-of-the-art systems with far less training data. The model trained almost entirely on synthetic videos. Its results add to the debate over whether video generat

中文介绍 Google DeepMind的研究指出，视频生成器可能已包含计算机视觉领域长期缺失的“世界模型”。其GenCeption项目将视频生成器重新应用于深度估计和分割等经典视觉任务，仅用少量训练数据就达到了与最先进系统相当的性能。该模型几乎完全基于合成视频进行训练，其成果为人工智能界关于世界模型和视觉理解的讨论提供了新视角。

bojieli/ai-agent-book

Python · ★ 5,259 · 🍴 495 · 📈 1,734 stars today

《深入理解 AI Agent：设计原理与工程实践》（李博杰 著）开源主仓库：全书正文、编译版 PDF 与按章配套代码

中文介绍 本项目是李博杰所著《深入理解 AI Agent：设计原理与工程实践》的开源主仓库。它提供了全书的完整正文、可编译的 PDF 版本，以及与各章节内容配套的代码示例。该项目旨在为读者提供一个全面的学习资源，深入探讨 AI Agent 的设计原理、架构模式及实际工程应用。开发者、研究人员或对 AI Agent 领域感兴趣的学习者，可以通过本书理论结合实践，掌握构建和部署智能 Agent 的核心知识与技能。

tirth8205/code-review-graph

Python · ★ 21,028 · 🍴 2,166 · 📈 551 stars today

Local-first code intelligence graph for MCP and CLI. Builds a persistent map of your codebase so AI coding tools read only what matters, with benchmarked context reductions on reviews and large-repo workflows.

中文介绍 Code Review Graph是一个本地优先的代码智能图谱工具，专为MCP（多组件项目）和CLI环境设计。它构建了一个持久化的代码库映射，旨在优化AI编码工具的上下文理解，确保AI在代码审查和处理大型仓库时仅读取关键信息。通过大幅减少AI上下文长度，提升了代码理解效率和审查质量。

kvcache-ai/ktransformers

Python · ★ 18,301 · 🍴 1,449 · 📈 328 stars today

A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations

中文介绍 ktransformers 是一个灵活的框架，专注于 LLM (大语言模型) 推理和微调的异构优化实践。该项目旨在帮助开发者和研究人员探索并应用多种优化技术，以提升 LLM 在不同硬件环境下的性能和效率。它简化了复杂的优化流程，使用户能够方便地试验和比较不同的优化策略，例如 KV cache 优化、量化、剪枝等，从而在部署和训练 LLM 时实现资源的高效利用和更快的响应速度。

rohitg00/ai-engineering-from-scratch

Python · ★ 39,588 · 🍴 6,614 · 📈 507 stars today

Learn it. Build it. Ship it for others.

中文介绍 ai-engineering-from-scratch 项目提供了一个从零开始学习 AI 工程的全面指南，涵盖 AI 系统从概念、开发、构建到部署的整个生命周期。旨在帮助开发者和工程师掌握构建、测试并交付可生产级 AI 应用所需的实践技能和最佳实践。它通过结构化的学习路径，适合希望系统性学习 AI 工程并提升实战能力的个人。

jamiepine/voicebox

TypeScript · ★ 43,247 · 🍴 5,286 · 📈 629 stars today

The open-source AI voice studio. Clone, dictate, create.

中文介绍 Voicebox 是一个开源的 AI 语音工作室，为用户提供强大的语音克隆、文本转语音及内容创作能力。它允许用户轻松克隆特定人声，将文本转化为逼真且情感丰富的语音，并基于此生成全新的音频内容。该项目旨在降低高质量语音合成技术的门槛，赋能内容创作者、开发者以及任何对个性化语音应用有需求的用户，广泛应用于播客制作、有声读物、虚拟助手或多媒体内容配音等场景。

KnockOutEZ/wigolo

TypeScript · ★ 1,749 · 🍴 108 · 📈 605 stars today

The go-to web for your AI coding agent — local-first search, fetch, crawl & research over MCP. No API keys, no cloud, $0/query. Public beta.

中文介绍 wigolo 为 AI 编程助手提供一个本地优先的信息检索与研究平台。它允许 AI 代理在无需 API 密钥、不依赖云服务且零成本的情况下，对本地代码项目 (MCP) 进行搜索、抓取和研究。这为开发者提供了一个私密、高效且完全离线的环境，用于增强 AI 编码助手的知识库，同时保护代码隐私并大幅降低运行成本。

andrewrabert/jellium-desktop

Rust · ★ 1,261 · 🍴 107 · 📈 54 stars today

An unofficial desktop client for Jellyfin

中文介绍 Jellium Desktop 是 Jellyfin 媒体服务器的一个非官方桌面客户端。Jellyfin 用户通常通过网页浏览器访问其媒体内容，而本项目则提供了一个独立的桌面应用程序。它旨在为用户提供更原生、更流畅的媒体浏览和管理体验，避免了浏览器标签页的干扰，并可能带来更好的性能表现。该客户端适用于那些希望将 Jellyfin 作为独立应用程序运行，以获得更沉浸式和便捷使用体验的用户。

github/copilot-sdk

Java · ★ 9,936 · 🍴 1,349 · 📈 46 stars today

Multi-platform SDK for integrating GitHub Copilot Agent into apps and services

中文介绍 GitHub Copilot SDK 是一个多平台开发工具包，旨在帮助开发者将 GitHub Copilot Agent 的强大 AI 辅助编程能力集成到自己的应用程序和服务中。通过此 SDK，各种开发环境、IDE 或自定义工具能够无缝接入 Copilot 的代码建议、自动补全等功能，从而为用户提供更智能、更高效的编程体验。它赋能开发者在更广泛的场景中利用 AI 提升生产力。

PostHog/posthog

Python · ★ 36,894 · 🍴 3,053 · 📈 424 stars today

🦔 PostHog is the leading platform for building self-driving products. Our developer tools – AI observability, analytics, session replay, flags, experiments, error tracking, logs, and more – capture all the context agents need to diagnose problems, uncover opportunities, and ship fixes. Steer it all

中文介绍 PostHog 是一个面向开发者的全栈产品分析与开发平台，旨在帮助团队构建“自驱动”产品。它提供了一系列工具，包括 AI 可观测性、用户行为分析、会话回放、功能开关 (feature flags)、A/B 实验、错误追踪和日志收集等。PostHog 帮助产品经理、开发者和数据分析师全面理解用户行为，快速迭代产品功能，并优化用户体验。

microsoft/terminal

C++ · ★ 104,137 · 🍴 9,436 · 📈 103 stars today

The new Windows Terminal and the original Windows console host, all in the same place!

中文介绍 Microsoft Terminal 项目包含了全新的 Windows Terminal 应用及其原始的 Windows 控制台主机组件。Windows Terminal 是一款现代化的、功能丰富的命令行界面，旨在为 Windows 用户提供优越的终端体验。它支持多标签页、GPU 加速文本渲染、自定义主题、以及集成 CMD、PowerShell 和 WSL 等多种 Shell 环境。该项目解决了传统 Windows 控制台体验的不足，为开发者和高级用户提供了高效、美观且高度可配置的命令行工具。

AstrBotDevs/AstrBot

Python · ★ 36,668 · 🍴 2,543 · 📈 62 stars today

AI Agent Assistant & development framework that integrates lots of IM platforms, LLMs, plugins and AI feature, and can be your openclaw alternative. ✨

中文介绍 AstrBot 是一个多功能的 AI Agent 助手和开发框架，旨在简化跨平台智能助手的构建与部署。它深度集成了多种即时通讯（IM）平台、主流大语言模型（LLM）以及丰富的插件和 AI 功能。该项目提供了强大的扩展性，允许开发者根据需求定制 Agent 的行为和能力，并可作为 OpenClaw 等工具的开源替代方案。AstrBot 适用于需要构建在不同社交和办公平台上线，并能灵活响应用户请求的智能 AI 助手的开发者。

1jehuang/jcode

Rust · ★ 8,829 · 🍴 1,024 · 📈 199 stars today

Coding Agent Harness

中文介绍 Jcode 是一个专为编码 Agent 设计的开发与评估框架（Coding Agent Harness）。该项目旨在为研究人员和开发者提供一套工具，用于构建、测试和比较 AI 编码助手的性能与能力。它能够帮助评估 AI 模型在代码生成、错误修复、重构以及其他编程任务中的表现。通过 Jcode，开发者可以更有效地迭代和优化 AI 驱动的编程工具，从而推动自动化编程和智能辅助编码技术的发展。

trycua/cua

HTML · ★ 20,211 · 🍴 1,341 · 📈 136 stars today

Scale computer-use 2.0 with open-source drivers, cross-OS fleets, and benchmarks for training, evaluation, and data generation.

中文介绍 CUA 旨在通过开源驱动、跨操作系统（OS）集群管理以及用于训练、评估和数据生成的基准测试，实现新一代的计算机自动化（"computer-use 2.0"）。该项目提供了一套工具和框架，支持开发者构建和部署能够与计算机进行交互的智能 Agent，无论它们运行在哪个操作系统上。它适用于需要大规模自动化操作、跨平台任务执行、合成数据生成或评估 AI Agent 在复杂计算环境中表现的开发者和研究人员。

MoonshotAI/kimi-cli

Python · ★ 9,837 · 🍴 1,192 · 📈 418 stars today

Kimi Code CLI is your next CLI agent.

中文介绍 Kimi Code CLI 是由 MoonshotAI 推出的一个命令行接口 (CLI) 工具，旨在作为开发者的“下一代 CLI 代理”。它将 AI 的能力直接带入终端，允许开发者通过命令行与 AI 进行交互，执行代码辅助、信息查询或自动化编程任务。Kimi-cli 适用于希望在命令行环境中高效利用 AI 能力、提升开发效率的程序员和工程师。

Flowseal/zapret-discord-youtube

Batchfile · ★ 31,017 · 🍴 2,374 · 📈 117 stars today

中文介绍 Flowseal/zapret-discord-youtube 项目，从其名称“zapret”（俄语意为“禁止”或“限制”）推断，可能是一个与 Discord 和 YouTube 平台上的内容限制或封锁相关的工具。它可能旨在帮助用户管理、识别或在特定情境下访问受限制的内容。具体功能可能包括处理被禁用的 YouTube 链接在 Discord 上的分享，或者提供绕过某些内容审查机制的解决方案。该项目适用于在这些平台上遇到内容访问障碍的用户或希望构建相关管理工具的开发者。

codecrafters-io/build-your-own-x

Markdown · ★ 528,855 · 🍴 50,032 · 📈 761 stars today

Master programming by recreating your favorite technologies from scratch.

中文介绍 build-your-own-x 提供了通过从零开始重新实现流行技术来掌握编程的实践路径。它鼓励学习者通过亲手构建诸如 Git、Docker、Redis 等系统，深入理解其底层原理和工作机制。该项目是为那些寻求通过项目实战巩固计算机科学基础、提升编程技能，并获得更深层次技术洞察力的开发者和自学者设计的。

lyogavin/airllm

Jupyter Notebook · ★ 23,596 · 🍴 2,678 · 📈 374 stars today

AirLLM 70B inference with single 4GB GPU

中文介绍 AirLLM 致力于解决大型语言模型 (LLM) 推理的硬件限制问题，使得 700 亿参数级别的模型能够在仅有 4GB 显存的单张 GPU 上高效运行。它通过优化模型加载、量化和计算方法，显著降低了 LLM 推理的资源门槛。这对于在本地设备、低成本硬件或边缘设备上部署复杂 AI 应用的开发者和研究人员极为有用。

Canner/WrenAI

Python · ★ 16,153 · 🍴 1,836 · 📈 96 stars today

GenBI (Generative BI) for AI agents, an open-source, governed text-to-SQL through an open context layer that turns natural-language questions into trusted dashboards, charts, and SQL across 20+ data sources, such as BigQuery, Snowflake, PostgreSQL, ClickHouse, Amazon Redshift, Databricks and more.

中文介绍 WrenAI 是一个为 AI Agent 设计的开源 GenBI (Generative BI) 平台，专注于实现受控的自然语言转 SQL (text-to-SQL) 功能。它通过一个开放的上下文层，能够将用户的自然语言问题转化为可信的 SQL 查询，并进一步生成交互式仪表板和图表。该项目支持连接超过 20 种不同的数据源，极大地简化了数据分析过程。WrenAI 适用于业务分析师、数据科学家和 AI Agent 开发者，帮助他们在无需深入 SQL 知识的情况下，快速从海量数据中获取洞察。

PKUFlyingPig/cs-self-learning

HTML · ★ 74,136 · 🍴 7,939 · 📈 128 stars today

计算机自学指南

中文介绍 cs-self-learning 是一个全面的计算机科学自学指南项目。它为希望自主学习或系统性提升计算机知识的人群，提供了一份结构化的学习路线和精选资源。该指南覆盖了从编程基础、数据结构与算法到操作系统、计算机网络、人工智能等多个核心领域。它旨在帮助学习者构建扎实的计算机科学理论基础和实践能力，是学生、转行者或任何对计算机科学充满热情者的宝贵学习资料。

You just hired a million bad employees.

@gsivulka · 10.2K 粉丝 · 591.2K 阅 · 512 赞 · 99 转

07/15 00:41

AI was supposed to replace human labor. It did the opposite. For the first time in history, humans are cheaper than software. And AI is creating more jobs than it eliminates. Technology has

中文介绍 博主gsivulka以「你刚雇佣了一百万个糟糕的员工」为引子，暗示了对AI工具或其应用效果的批判性观点。

Making Fable Cheaper Than Opus

@joon_h_lee · 515 粉丝 · 366.3K 阅 · 521 赞 · 43 转

07/14 01:03

We replaced Opus 4.8 with Fable 5, and Devin’s bill went down. Fable 5 costs twice as much per token as Opus 4.8. But when we ran both models on FrontierCode 1.1 using our new Fusion architecture,

中文介绍 博主分享了通过用Fable 5替换Opus 4.8显著降低Devin账单的经验。尽管Fable 5的每token成本是Opus 4.8的两倍，但利用其新Fusion架构在FrontierCode 1.1上运行模型后，整体费用反而下降。这表明通过优化模型架构和运行时环境，即使使用成本更高的模型也能实现更优的经济效益。

The Most Human Technology Ever Made

@illscience · 30.4K 粉丝 · 332.5K 阅 · 539 赞 · 69 转

07/14 01:05

Over the weekend, Mira Murati wrote a phenomenal essay that you should read in full: “The future worth building is human.” I think it’s one of the best essays of the year, because it gets at the

中文介绍 博主推荐OpenAI首席技术官Mira Murati撰写的文章「值得建设的未来是人性化的」。该文章探讨了将人类价值融入AI发展的未来方向，并被博主评价为年度最佳文章之一。它强调了在AGI时代，技术应以人为本，而非单纯追求技术突破。

GLM 5.2 With Vision

@part_harry_ · 1.5K 粉丝 · 225.6K 阅 · 503 赞 · 46 转

07/16 12:24

GLM 5.2 is one of the best currently available open source language models. However, unlike other flagship models like Qwen, Kimi and Minimax, GLM 5.2 does not support image inputs. We took this as a

中文介绍 博主讨论开源语言模型GLM 5.2的视觉能力（Vision），可能是在为其探索或开发图像输入功能。这一举措旨在增强GLM 5.2的多模态处理能力，使其能够处理视觉信息，从而拓宽应用范围并提升其在与Qwen、Kimi等模型竞争时的竞争力。

The Self-Driving Company

@amasad · 472.6K 粉丝 · 223.7K 阅 · 560 赞 · 65 转

07/17 01:07

We are beginning to see what happens when a company learns to operate itself. In the past six months, engineers at Replit have nearly tripled code output. Review times held steady. Reversions and

中文介绍 博主提出了「自驱动公司」这一概念，暗示企业运营未来将高度自动化，能像自动驾驶汽车一样自我管理和优化。这可能涉及利用AI和自动化技术，大幅提升公司的运营效率、决策速度和资源配置能力，预示着企业组织形态和生产力的一次重大变革。

Maybe Intelligence Ain't All That

@CliffordSosin · 16.4K 粉丝 · 197.0K 阅 · 514 赞 · 49 转

07/19 05:36

Superintelligence arrived. You probably didn't notice, because it turned out to be kind of incremental. Don't believe me? Run the test. Talk to Fable 5 for an hour, then talk to your ten smartest

中文介绍 博主探讨了「超级智能」的到来，指出其以渐进式而非颠覆性方式出现，因此可能未被察觉。他建议进行一项测试：与 Fable 5 交流一小时，再与十位最聪明的人交流，以感受这种「增量式超级智能」。

How OpenAI’s Sol Finally Learned Design Taste

@DesignArena · 17.6K 粉丝 · 191.5K 阅 · 509 赞 · 46 转

07/16 00:36

We benchmarked GPT-5.6 Sol on Design Arena’s Web Design (Non-Agentic) Arena, and we were surprised to find that it ranks 1st overall. This is 18 places higher than its predecessor GPT-5.5, and is the

中文介绍 DesignArena 提出了「How OpenAI’s Sol Finally Learned Design Taste」的问题，可能旨在探讨 OpenAI 的 Sol 模型在学习和掌握设计美学方面的进展和方法。

How to build your first team of agents:

@EXM7777 · 126.7K 粉丝 · 173.6K 阅 · 521 赞 · 51 转

07/18 21:48

I'm going to show you how to build your first team of AI agents... and how to wire them into a self improving loop inside a shared workspace the team we're building has 5 agents, one for each function

中文介绍 该推文详细讲解如何构建首个 AI 智能体团队，并将其整合到自我改进循环的工作流中。教程涵盖在共享工作空间内连接智能体的步骤，特别提到构建一个由 5 个具备独立功能的智能体组成的团队。

The Next Horizon in Agents

@btaylor · 158.6K 粉丝 · 144.3K 阅 · 511 赞 · 40 转

07/16 22:48

Introducing Horizon Today, Sierra is announcing Horizon, a platform that enables agents to pursue long-horizon goals, like originating a loan or getting prior authorization for a healthcare procedure.

中文介绍 Sierra公司发布了名为Horizon的新平台，旨在赋能AI Agent处理复杂且「长周期」的实际业务目标。该平台支持Agent完成如发起贷款或获取医疗程序预授权等多步骤任务。这标志着AI Agent从执行短期指令向处理更深层次、更具商业价值的复杂流程迈进，为企业自动化提供了新的可能性。

Gemini API Managed Agents: free tier, triggers & more

@GoogleAIStudio · 186.6K 粉丝 · 111.0K 阅 · 512 赞 · 51 转

07/17 01:05

Today we’re announcing new capabilities for Managed Agents in Gemini API, including free tier availability, budget control guardrails, and scheduled triggers. Building on our previous release of

中文介绍 Google AI Studio宣布Gemini API的Managed Agents新增功能，包括提供免费层级、预算控制保护措施以及支持定时触发器。这些更新建立在之前的发布基础上，旨在降低开发者使用门槛，赋予他们更大的灵活性来管理和部署AI Agent。新功能将促进更广泛的Agent应用，并帮助用户更好地控制成本与调度。

A Framework for Frontier AI and the Dawning of a New Age

@demishassabis · 1.4M 粉丝 · 98.1K 阅 · 1.5K 赞 · 236 转

07/14 17:10

This is a pivotal moment in human history. Artificial General Intelligence (AGI), a system that exhibits all the cognitive capabilities the brain has, is probably only a few short years away. When we

中文介绍 Google AI CEO Demis Hassabis认为，人类正处于关键历史时刻，通用人工智能（AGI）的实现可能仅在数年之内。AGI系统有望展现人类大脑的全部认知能力。他可能在分享一个关于前沿AI的框架，以应对即将到来的新时代。

How we built our knowledge base

@cerebras · 60.4K 粉丝 · 70.7K 阅 · 509 赞 · 38 转

07/17 02:27

Authors: @hi_im_isaac_, @learnwdaniel, @gaozenghao note: the interactive version of full technical blog available: https://www.cerebras.ai/blog/how-we-built-our-knowledge-base Employees ask our internal knowledge base more than 15,000

中文介绍 Cerebras发布技术博客，详细阐述了他们如何构建企业内部知识库。该知识库每月处理员工超过15,000次的查询，极大提高了工作效率。博文由多位作者共同撰写，提供了实际案例和技术细节，为其他组织构建高效智能知识管理系统提供了宝贵的实践经验和设计思路。

Kimi K3 on KernelBench

@elliotarledge · 40.7K 粉丝 · 69.3K 阅 · 506 赞 · 45 转

07/17 17:24

This is the Kimi K3 post you guys have been waiting for. I got some early access to this model and have been testing it on kernels, and even before seeing the benchmark scores I was impressed with its

中文介绍 博主获得了Kimi K3模型的早期访问权限，并对其进行了初步测试和KernelBench基准评估。在未查看具体分数前，博主已对Kimi K3的性能印象深刻。这篇分享为期待Kimi K3的用户提供了第一手体验和性能洞察，预示着该模型可能在某些任务上展现出强大的能力。

I Brutally Modified My Front-End Design Skill ~ Now My UIs Don’t Look Like AI Crap

@JoePro · 949 粉丝 · 66.8K 阅 · 537 赞 · 29 转

07/14 11:51

To create a custom Claude skill, navigate to Customize > Skills in your Claude interface, click the + button, and select + Create skill

中文介绍 详细介绍了在 Claude 界面中创建自定义 skill 的具体步骤。用户需进入「Customize > Skills」，点击「+」按钮，然后选择「+ Create skill」即可。这为希望扩展 Claude 功能的用户提供了清晰的实践指导。

How I'd make $10 million with AI agents

@gregisenberg · 682.7K 粉丝 · 62.2K 阅 · 578 赞 · 36 转

07/14 02:22

2 markets worth trillions are opening at the same time. I've been through 3 platform shifts living in Silicon Valley building/selling VC-backed companies and I've NEVER seen this. Mobile apps came

中文介绍 博主分享了对当前 AI 代理市场机遇的独到见解。他指出有「2个万亿级市场」正同步涌现，并将其与过往的「移动应用」等重大平台转型相提并论，认为这是前所未有的巨大商业机会。

2 Hermes Workflows I can't live without

@0xJeff · 80.8K 粉丝 · 61.6K 阅 · 502 赞 · 36 转

07/13 19:33

I’ve been consistently using Hermes everyday for the past 4 months and he’s already a core part of my research & investment workflows. Before Hermes, it was just me relying on Grok, Claude, Gemini for

中文介绍 博主分享了他持续使用 Hermes 4个月后，认为不可或缺的「2个工作流」。这些工作流已成为其研究与投资流程的核心组成部分，相较于此前单独依赖 Grok、Claude、Gemini 等工具，显著提升了效率与体验。

Let's build Claude Code's harness (step-by-step)

@akshay_pachaar · 281.7K 粉丝 · 53.7K 阅 · 516 赞 · 74 转

07/16 02:10

We will cover everything that goes into building a coding harness, the agent loop, planning, subagents, sandboxing, memory, and checkpointing, built step by step. If you've ever tried building your

中文介绍 博主分享了为 Claude Code 构建编码辅助工具「coding harness」的逐步教程。内容深入讲解了智能体循环、规划、子智能体、沙盒环境、内存管理和检查点等关键技术组件的实现细节。

From Loop Engineering to Graph Engineering?

@IntuitMachine · 62.3K 粉丝 · 50.7K 阅 · 552 赞 · 47 转

07/18 18:00

What the shift in AI agent architecture is really about Peter Steinberger just posted nine words that gathered thousands of likes: https://x.com/steipete/status/2078277297791189132 "Are we still talking loops or did we

中文介绍 该推文探讨了 AI 智能体架构的关键转变，从传统的「循环工程」（Loop Engineering）转向「图工程」（Graph Engineering）。文章引用 Peter Steinberger 的热门观点，提出对当前智能体设计范式的深刻反思。

Hierarchical Denoising For Multi-Step Visual Reasoning

👍 3

07/16 08:00

Video models are evolving into vision foundation models, yet they still lack human-like multi-step reasoning. Streaming autoregressive diffusion models are efficient but limited in reasoning, while bidirectional diffusion enables global revision with high inference costs due to dense frame-level den

中文介绍 视频模型正发展为视觉基础模型，但在多步视觉推理方面仍欠缺类人能力。该研究探讨了流式自回归扩散模型和双向扩散模型的优缺点，前者高效但推理有限，后者支持全局修正但推理成本高昂。论文提出一种分层去噪方法，旨在提升视频模型的复杂视觉推理能力，以克服现有模型的局限性。

Partition, Prompt, Aggregate: Statistical Self-Consistency in Language Models

👍 7

07/16 08:00

In-context learning is commonly interpreted as a form of conditional inference, in which the prompt specifies a context and the model's output is treated as an estimate of the corresponding conditional distribution. If this interpretation holds, then LLM estimates should satisfy basic probabilistic

中文介绍 上下文学习常被视为一种条件推理形式，其中提示词设定语境，模型输出则被视为对应条件分布的估计。该论文提出「Partition, Prompt, Aggregate」方法，旨在探讨并提升大型语言模型（LLMs）估计的统计自洽性。研究表明，如果这种解释成立，LLM的估计应满足特定的统计特性，从而提高模型输出的可靠性。

RoboTTT: Context Scaling for Robot Policies

👍 18

07/16 08:00

Recent robot foundation models operate with single-step or short-history visuomotor context. We introduce Test-Time-Training Robot Policies (RoboTTT), a robot model and training recipe that scale visuomotor context to 8K timesteps, three orders of magnitude beyond state-of-the-art policies, without

中文介绍 现有的机器人基础模型通常仅处理单步或短期视觉运动上下文。该论文提出「RoboTTT」（测试时训练机器人策略），这是一种机器人模型和训练方法，能将视觉运动上下文扩展到8K时间步。这比现有技术水平提升了三个数量级，有望显著增强机器人策略的长期规划和决策能力。

MeanFlowNFT: Bringing Forward-Process RL to Average-Velocity Generators

👍 12

07/16 08:00

MeanFlow generators achieve fast few-step sampling by predicting average velocities over time intervals, making them attractive for efficient generation. Reinforcement learning (RL) has become a powerful way to align diffusion and flow models with human preferences and task-specific objectives. In p

中文介绍 MeanFlow生成器通过预测时间间隔内的平均速度，实现快速的少步采样，适用于高效生成任务。强化学习（RL）已成为使扩散模型和流模型与人类偏好及特定任务对齐的有效方法。该论文提出了「MeanFlowNFT」，将前向过程强化学习引入平均速度生成器，旨在进一步优化生成模型的性能和用户体验。

SearchOS-V1: Towards Robust Open-Domain Information-Seeking Agent Collaboration

👍 60

07/16 08:00

Recent advances in Tool-Integrated Large Language Models have made web search a core capability of information-seeking agents. However, as interaction histories grow, agents increasingly struggle to track task progress. When search attempts fail to yield useful evidence, current single- and multi-ag

中文介绍 随着工具集成大型语言模型的发展，网络搜索已成为信息搜寻代理的核心能力。然而，交互历史的增长导致代理难以跟踪任务进度，尤其当搜索未能提供有效证据时。该论文提出「SearchOS-V1」，旨在构建更鲁棒的开放领域信息搜寻代理协作系统，以解决现有代理在处理复杂和失败搜索场景中的局限性。

HoloGeo: Mitigating Landmark Bias in Geo-localization via Evidence-Driven Reasoning

👍 0

07/16 08:00

Recent advances in Vision-Language Models (VLMs) have significantly improved image geo-localization, yet existing models remain susceptible to landmark bias, causing them to overlook geographical cues or form spurious correlations, ultimately resulting in inaccurate localization. To systematically i

中文介绍 视觉语言模型（VLMs）显著提升了图像地理定位能力，但现有模型易受地标偏见影响，导致忽视地理线索或形成虚假关联，最终定位不准确。该论文提出「HoloGeo」方法，旨在通过证据驱动推理减轻这种地标偏见。通过更有效地整合视觉与语言信息，HoloGeo有望提升地理定位的鲁棒性和准确性。

BadWAM: When World-Action Models Dream Right but Act Wrong

👍 46

07/16 08:00

World-action models (WAMs) are emerging as a promising foundation for embodied control: rather than predicting actions alone, they learn representations that couple action generation with future world prediction. This coupling is often viewed as a source of robustness, interpretability, and safety,

中文介绍 世界-行动模型（WAMs）通过将行动生成与未来世界预测相结合，成为具身控制的潜在基础。这种耦合通常被视为鲁棒性和可解释性的来源。然而，该论文提出的「BadWAM」研究了WAMs在“正确预测世界却错误执行行动”的情况。文章分析了导致这种偏差的原因，并探讨了提升模型行动准确性的策略。

MM-IssueLoc: A Controlled Benchmark for Evaluating Visual Evidence in Multimodal Repository-Level Issue Localization

👍 0

07/16 08:00

Real repository issues routinely include visual evidence such as screenshots, error dialogs, rendered UI states, and logs, yet repository-level issue localization is evaluated mostly as a text-only task. Existing multimodal SE benchmarks evaluate end-to-end repair, entangling localization with patch

中文介绍 实际软件仓库问题常包含截图、错误对话框等视觉证据，但仓库级问题定位主要仍是文本任务。现有多模态软件工程基准侧重端到端修复，而非纯粹定位。该论文提出「MM-IssueLoc」，一个受控基准测试集，用于专门评估多模态仓库级问题定位中视觉证据的作用。它旨在推动更准确、更全面的问题分析。

Setup Complete, Now You Are Compromised: Weaponizing Setup Instructions Against AI Coding Agents

👍 0

07/16 08:00

AI coding agents set up projects by reading documentation and installing the dependencies it lists, without verifying their names, sources, or known vulnerabilities. By editing only a README, requirements file, or Makefile, an attacker can redirect the agent to an untrusted registry, a known-vulnera

中文介绍 AI编程代理通过阅读文档安装依赖项，但缺乏对名称、来源或已知漏洞的验证。攻击者可通过修改README文件、依赖列表或Makefile，将代理重定向到不受信任的注册表，从而导致系统被入侵。该论文揭示了针对AI编程代理的潜在供应链攻击风险，并强调了加强其安装过程安全性的必要性。

Long-Context Fine-Tuning with Limited VRAM

👍 0

07/16 08:00

Parameter-efficient fine-tuning reduces model and optimizer memory, but dense attention still makes long training sequences expensive. We combine Hierarchical Global Attention (HGA) with segment-wise backpropagation and tiered KV storage. Only the active segment remains differentiable in VRAM; older

中文介绍 一篇HuggingFace论文旨在解决显存受限条件下进行长上下文微调的挑战。现有参数高效微调虽能降低模型和优化器内存，但密集注意力机制仍使长序列训练成本高昂。研究提出一种新方法，结合了分层全局注意力（HGA）与分段反向传播及分层KV存储技术，以有效降低长序列训练的开销，尤其适用于VRAM有限环境下的长上下文微调。

SUFLECA: Scaling Up Feature Learning for CAD-to-image Alignment

👍 7

07/16 08:00

CAD-to-image alignment aims to estimate an object's 9D pose (rotation, translation, and anisotropic scale) from a single RGB image, enabling applications in robotics and augmented reality. Recent zero-shot methods use visual foundation models to match image regions to CAD models, yet typically their

中文介绍 CAD-到图像对齐技术旨在从单张RGB图像中估计物体9D姿态（旋转、平移和各向异性尺度），在机器人和增强现实中具有应用。近期零样本方法利用视觉基础模型进行图像区域与CAD模型匹配。该论文提出「SUFLECA」方法，旨在扩展特征学习规模，以提升CAD-到图像对齐任务的准确性和鲁棒性。

Video = World + Event Stream

👍 17

07/16 08:00

We present Wan-Streamer v0.3, which reframes our native-streaming interaction model under a single organizing view: a video is a world plus an event stream. The world is the persistent context in which a video unfolds, including the environment, scene, subjects, ambient acoustic conditions, voice ch

中文介绍 该论文推出「Wan-Streamer v0.3」，提出一种将视频重新定义为“世界+事件流”的交互模型。其中，“世界”代表视频展开的持久性上下文，涵盖环境、场景、主体和环境声学等；“事件流”则捕捉动态变化。这一新框架旨在为本地流媒体交互提供统一视图，深化对视频内容的理解和处理。

AeroAct: Action-Centered World-Action Models for Language-Conditioned Quadrotor Flight

👍 0

07/16 08:00

Language-conditioned quadrotor flight requires a policy to ground semantic goals, anticipate the visual consequences of ego-motion, and output control references that remain smooth and dynamically executable under rapidly changing first-person views. Existing aerial vision-language navigation and vi

OmniaBench: Benchmarking General AI Agents Across Diverse Scenarios

👍 0

07/16 08:00

Large language models are increasingly evolving from text generators into general agents capable of understanding user requests, invoking external tools, and completing complex tasks through interaction. However, existing agent benchmarks often focus on limited scenarios, tool ecosystems, or interac

Demographically-Conditioned Synthetic Medical Images for Bias Mitigation and Bias Detection in Disease Classifiers

👍 0

07/16 08:00

Per-subgroup fairness audits of medical image classifiers face a sample-size problem: minority subgroups in held-out test sets have so few samples that the resulting confidence intervals on per-subgroup performance are wider than the bias the audit is meant to detect. We argue that a demographically

From Draft to Draft-Free: One-Step Video Object Removal via Privileged Distillation and Fast Planting

👍 0

07/16 08:00

Video object removal is a fundamental yet challenging task in video editing. Despite recent progress, existing methods typically fall into two categories. Traditional approaches based on optical flow or attention mechanisms often introduce noticeable artifacts and yield unnatural results. In contras

LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget

👍 175

07/16 08:00

A growing gap separates inference context lengths from RL post-training: inference systems are approaching million-token contexts, while post-training workloads often remain at 256K tokens or below and rely on length generalization at deployment. The gap is especially important for AI agents, whose

Steering Robustness into World Action Models via Mechanistic Interpretability and Optimal Control

👍 0

07/16 08:00

World Action Models (WAMs) enable semantically- and physically-informed control but are brittle under distribution shift. In this work, we use mechanistic interpretability to study how robustness-relevant perturbations are represented in WAM activation space. Comparing activations across successful

VideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding

👍 145

07/16 08:00

Recent advances in video understanding have spanned motion, long video, and streaming interaction, driving this field toward real-world applications. Despite this progress, current open-source models remain limited in several ways. They often struggle to generalize across diverse video types, making

FlashDecoder: Real-Time Latent-to-Pixel Streaming Decoder with Transformers

👍 0

07/16 08:00

Real-time video generation demands fast decoding as much as fast denoising, yet current latent video diffusion models rely on 3D convolutional decoders that are slow and memory-intensive at high resolutions or for long video. We introduce FlashDecoder, a fast, memory-efficient pure-Transformer video

StructureClaw: Traceable LLM Agents and an Executable Benchmark for Structural Engineering Workflows

👍 0

07/16 08:00

Addressing a structural-engineering request requires more than a single answer; it requires a chain of interdependent artifacts: interpreted requirements, a computable model, validation records, solver outputs, code-check records, and a final report. Evaluations centered on question answering or scr

Proof-or-Stop: Don't Trust the Agent, Trust the Evidence -- Loop Engineering for Verifiable Evidence-Gated Lifecycle Control

👍 0

07/16 08:00

Autonomous coding agents increasingly execute multi-step software work, but lifecycle states such as reviewed, tested, DONE, and ready-to-merge remain claims unless supported by current evidence. We present Proof-or-Stop Lifecycle Control, a method that permits lifecycle transitions only when fresh,

Towards Human-like Physical Intelligence: LifelongVision-Language-Action Learning for Robotic Manipulation

👍 0

07/16 08:00

Similar to the natural capabilities of humans to sequentially learn new tasks, robots with Vision-Language-Action (VLA) models should possess lifelong learning ability to learn a new task when deployed in open-world environments. However, most recently proposed lifelong learning models aim to effect

SEED: Self-Evolving On-Policy Distillation for Agentic Reinforcement Learning

👍 85

07/16 08:00

Large language models are increasingly trained as interactive agents for long-horizon tasks involving multi-turn interaction, tool use, and environment feedback. Outcome-based reinforcement learning (RL) provides a practical optimization paradigm, but its sparse trajectory-level rewards offer limite

WanSong v1.0 Technical Report

👍 13

07/16 08:00

Music generation foundation models have recently attracted significant industry attention. However, achieving efficient generation and high-fidelity long-form audio while supporting controllability remains challenging. To address these needs, we present WanSong, a simple yet powerful approach for lo

Does Multi-Agent Debate Improve AI Feedback on Research Papers?

👍 0

07/16 08:00

Probably not, at least for meta-analyses in economics. In a pre-registered, identity-masked, within-paper experiment, the authors of 44 meta-analyses ranked three AI reports on their own paper by usefulness for improving it: a single pass by a frontier model against two multi-agent debate tools we b

Reflex: Real-Time VLA Control through Streaming Inference

👍 0

07/16 08:00

Flow matching Vision-Language-Action (VLA) models promise precise continuous control, but their iterative denoising nature introduces fundamental incompatibilities with real-time robotics: global timestep injection invalidates KV-caching, forcing a choice between slow O(N^2) re-computation or mathem

Project Kaleidoscope: Contextual, Human-Aligned Evaluation for Real-World AI Applications

👍 0

07/16 08:00

Evaluations (Evals) are a deployment bottleneck for real-world AI applications: public benchmarks rarely match a team's users, context, or policies, and human review is often tedious to scale. Motivated by our work with AI applications in the public sector, this project addresses recurring evaluatio

VIABench: A Comprehensive Video Benchmark Collected from Blind Individuals for Visual Impairment Assistance

👍 8

07/16 08:00

Visually impaired individuals (VIIs) encounter significant daily challenges due to limited access to visual information. Although Multimodal Large Language Models (MLLMs) have achieved impressive results on general vision and language tasks, their practical utility in real-world blind assistance sti

MCPEvol-Bench: Benchmarking LLM Agent Performance Across Dynamic Evolutions of MCP Servers

👍 0

07/16 08:00

As Model Context Protocol (MCP) servers emerge as the core infrastructure for connecting LLMs with external tools, existing benchmarks leverage real-world MCP servers to evaluate LLM agents' tool-using capabilities. However, these benchmarks overlook the continuous evolution of tool interfaces and f

Kobbe

07/20 04:35

Privacy-friendly web analytics with traffic and revenue.

中文介绍 Kobbe是一款注重用户隐私的网络分析工具，旨在帮助网站拥有者追踪流量和收入数据，同时保护用户隐私，提供友好的隐私管理功能。

Spycost

07/20 04:34

Fell for a discount again?

中文介绍 Spycost旨在帮助消费者避免因折扣诱惑而冲动购物。其具体功能和运作方式在现有信息中未详细说明，但预期是提供消费辅助以避免非理性购买。

OpenSEO

07/20 04:34

The open source Ahrefs alternative

中文介绍 OpenSEO是一个开源的搜索引擎优化（SEO）工具，旨在为用户提供Ahrefs的替代方案。它帮助网站提升搜索排名和流量，提供透明和可定制的SEO解决方案。

Rewisp

07/20 04:28

See it once. Ask forever.

中文介绍 Rewisp是一款专为Mac用户设计的环境记忆工具。它旨在帮助用户记录和回忆浏览过的信息，从而实现「看一次，永远可问」的功能，提升信息管理效率。

BaseRT

07/20 04:27

6.4x faster than llama.cpp, 3.9x faster than MLX

中文介绍 BaseRT是一个高性能的计算框架或推理引擎，其运行速度比llama.cpp快6.4倍，比MLX快3.9倍。这显示其在处理特定任务上具有显著的性能优势和效率。

Detourmap

07/20 04:16

Every place worth the detour, on one interactive map

中文介绍 Detourmap是一个交互式地图应用，它汇集了各种值得「绕道前往」的独特地点。用户可以通过该地图发现和探索那些非传统、具有特别体验的场所。

Mainichi

07/20 04:11

Learn Japanese, one prefecture at a time.

中文介绍 Mainichi 是一款日语学习应用，其特色是每次专注于学习一个日本行政区（都道府县）的知识，帮助用户逐步掌握日语和日本文化。

Clark

07/20 04:10

An AI coworker with its own cloud computer

中文介绍 Clark 是一款具有自身云计算机的 AI 协作助手，旨在作为人工智能同事，帮助用户处理日常工作任务，提升效率。

Mirage

07/20 04:10

Turn your SaaS into a clickable demo in 90 seconds.

中文介绍 Mirage 是一款可将 SaaS 产品快速转换为可点击演示的工具。用户能在 90 秒内创建出交互式产品演示，便于潜在客户体验和了解产品功能。

ZooData

07/20 04:10

The data layer for AI agents

中文介绍 ZooData 是一款专为 AI 代理设计的数据层产品，旨在为人工智能代理提供高效的数据处理和管理能力，支持其更智能地运行。

Devin AI: The Full Beginner’s Guide (Better Than Claude Code?)

07/17 02:25

中文介绍 这则YouTube视频提供了Devin AI的全面初学者指南，深入探讨了这款人工智能工具在代码开发方面的应用潜力。视频可能通过演示和比较，向观众展示Devin AI的功能和使用方法，并讨论其与Claude Code相比的优劣，以帮助初学者快速掌握。

Director by OpenArt

07/17 02:12

中文介绍 这则YouTube短视频介绍了OpenArt公司推出的一款名为「Director」的新产品或功能。该视频可能简要展示了「Director」的主要特点和用途，暗示其可能是一款用于创意或艺术领域的人工智能工具。

Codex and Claude Shipped Browser Updates. This Changes Everything.

07/15 21:00

中文介绍 Codex和Claude两大AI模型近期发布了浏览器更新，预计将对用户与AI的互动方式产生显著影响。这些更新可能包括功能改进、性能优化或更深度的浏览器集成，旨在提升用户体验并拓宽AI工具的应用场景。

OpenAI Just Merged ChatGPT and Codex. This Changes Everything.

07/13 01:24

Grok 4.5 + Cursor: Full Guide

07/10 01:51

Build Amazing Websites with Claude Fable 5 (Full Guide)

07/07 01:37

How to Create Your Own AI Assistant (It's So Easy)

07/01 09:40

Poweful Agents inside Slack

06/25 06:39

AI Agents Just Changed Forever: GLM 5.2, Codex Skills, Claude & Cursor

06/21 21:30

9 AI Agent Skills To Get Ahead of 99% of People

06/18 10:27

为什么“节省 90% Token”不等于 Coding Agent 总成本降低 90%

07/19 21:12

6 回复 · 程序员 节点

Vibe Coding 副业四个月：成本超出收益，要放弃吗？

07/19 19:17

31 回复 · 程序员 节点

Kimi K3 这波真的爆了，你们用着感觉怎么样？

07/19 18:11

25 回复 · 程序员 节点

iOS 如何设置能实现像安卓手机的来电提醒一样显示是谁打电话过来的，比如 10086 标识为中国移动这样？

07/19 11:28

11 回复 · Apple 节点

iPhone 换电池，都是去官方还是自己买电池换呢？

07/18 20:19

15 回复 · Apple 节点

强烈推荐一款性价比极高的 Linux 本搞 vibe coding

07/17 12:24

31 回复 · Linux 节点

国行 iPhone 开始出现苹果 AI 流量连接

07/17 11:51

16 回复 · Apple 节点

如果以后 Apple AI 国行版上线，国行的 iPhone 可以在科学上网前提下，使用国外的 Apple AI 吗？

07/16 16:24

16 回复 · Apple 节点

已故妻子的 iPhone 因失窃设备保护卡在 Face ID 验证，已知锁屏及 Apple 账户密码，能否提取微信数据？

07/16 16:07

103 回复 · Apple 节点

sing box 有望回归 apple store

07/16 15:11

12 回复 · Apple 节点

Heavy TV watching associated with smaller brain structures, study finds

07/20 04:25

3 points · 0 comments

Modder Runs GTA III Inside GTA: San Andreas on an In-Game TV

07/20 03:27

14 points · 2 comments

Orion Browser by Kagi

07/20 03:13

8 points · 5 comments

HomeLab #1: MikroTik as a Home Router

07/20 02:57

14 points · 6 comments

Neither GCC nor Clang are compliant with standard C++

07/20 01:32

28 points · 25 comments

HMD Touch 4G

07/20 01:25

49 points · 42 comments

The Last MPEG-4 Visual Patent Has Expired

07/20 00:45

86 points · 17 comments

Moonshot AI suspends new subscriptions due to Kimi K3 demand

07/20 00:02

120 points · 44 comments

C64 Basic Dungeon Crawler: Goblin Attack (C64 Basic Part 8)

07/19 23:20

40 points · 1 comments

Natural experiments prove phytoplankton carbon removal works

07/19 22:51

14 points · 6 comments

Show HN: I replaced a $120k bowling center system with $1,600 in ESP32s

07/19 22:41

I might be the only SRE on Earth with his own bowling center. It's a more in-depth gig than you'd think.My family and I bought an abandoned 8-lane bowling center in the rural mid-west. In our small town there weren't many recreation options for families. You've heard of a food de

Bananas sprout in Rayleigh Garden UK after 15 years

07/19 21:29

84 points · 63 comments

Minecraft: Java Edition now uses SDL3

07/19 19:48

213 points · 144 comments

I joined the IndieWeb, here's what I learned

07/19 19:14

108 points · 70 comments

The death and rebirth of my home server

07/19 18:44

97 points · 67 comments

What I learned selling 2,500 MIDI recorders: Hardware is not so hard

07/19 18:34

352 points · 165 comments

Claude Code uses Bun written in Rust now

07/19 18:03

321 points · 426 comments

Qwen 3.8

07/19 16:44

https://www.qwencloud.com/pricing/token-plan

OpenAI reduces Codex Model Context Size from 372k to 272k

07/19 15:54

259 points · 118 comments

Transcribe.cpp

07/19 08:38

696 points · 144 comments

Codex Resets

07/19 07:24

272 points · 175 comments

UnifiedIR for Julia

07/19 05:28

61 points · 15 comments

Building an Arch Linux Aarch64 Port for Holo Core

07/17 23:50

15 points · 0 comments

Speech Recognition and TTS in less than 500kb

07/15 03:25

544 points · 79 comments

Blender 5.2 LTS

07/15 02:22

286 points · 113 comments

The Zen of Parallel Programming

07/14 22:23

25 points · 0 comments

Infinities, impossibilities, and the man in the white linen suit

07/14 16:24

49 points · 38 comments

Mathematicians still don't know the fastest way to multiply numbers

07/14 01:21

200 points · 112 comments

Land Atlas – soil, farmability, and crop analysis for land listings

07/13 19:58

39 points · 12 comments

Cagire: Live Coding in Forth

07/10 22:45

59 points · 9 comments

v2.1.215

07/19 10:56

What's changed Claude no longer runs the /verify and /code-review skills on its own; invoke them with /verify or /code-review when you want them

中文介绍 Anthropic公司旗下Claude代码工具发布v2.1.215版本更新。此次更新调整了"/verify"和"/code-review"技能的运行方式，它们不再自动执行，用户需要时需通过指令明确调用。

v2.1.214

07/18 09:20

What's changed Fixed single-segment dir/** allow rules like Edit(src/**) auto-approving writes to nested dir/ directories anywhere in the tree instead of only /dir Fixed a permission-check bypass affecting commands run in Windows PowerShell 5.1 sessions Fixed Bash permission checks to fail closed on

中文介绍 Anthropic Claude Code发布了v2.1.214版本。此更新主要修复了多项错误，包括单段目录通配符允许规则、影响Windows PowerShell 5.1会话的权限检查绕过问题，以及Bash权限检查错误。

v2.1.212

07/17 08:26

What's changed /fork now copies your conversation into a new background session (its own row in claude agents) while you keep working; the in-session subagent it used to launch is now /subtask Added claude auto-mode reset to restore the default auto-mode configuration, with a confirmation prompt (pa

中文介绍 Anthropic 的 Claude Code 项目发布了 v2.1.212 版本。主要更新包括："/fork" 命令现在会将当前对话复制到一个新的后台会话中，用户可继续工作；原有的会话内子代理功能则被新的 "/subtask" 命令取代。此外，还新增了 "claude auto-mode reset" 功能，用于恢复默认的自动模式配置。

v2.1.211

07/16 07:02

What's changed Added --forward-subagent-text flag and CLAUDE_CODE_FORWARD_SUBAGENT_TEXT environment variable to include subagent text and thinking in stream-json output Fixed permission previews relayed to chat channels not neutralizing bidirectional-override, zero-width, and look-alike quote charac

v2.1.210

07/15 07:45

What's changed Added a live elapsed-time counter to the collapsed tool summary line so long-running tool calls visibly tick instead of looking stuck Added a startup warning for Write(path), NotebookEdit(path), and Glob(path) permission rules — use Edit(path) or Read(path) instead Fixed isolation: 'w

v2.1.209

07/14 14:36

What's changed Fixed /model and other dialogs being blocked in claude agents background sessions (reverts an overly broad guard)

v2.1.208

07/14 09:10

What's changed Added screen reader mode: opt-in plain-text rendering for screen reader users. Run claude --ax-screen-reader, set CLAUDE_AX_SCREEN_READER=1, or add "axScreenReader": true to settings. Added vimInsertModeRemaps setting: map two-key insert-mode sequences like jj to Escape in vim mode Ad

v2.1.207

07/11 08:52

What's changed Auto mode is now available without CLAUDE_CODE_ENABLE_AUTO_MODE opt-in on Bedrock, Vertex AI, and Foundry; disable via disableAutoMode in settings Fixed the terminal freezing and keystrokes lagging while streaming responses containing very long lists, tables, paragraphs, or code block

v2.1.206

07/10 09:45

What's changed Added directory path suggestions to /cd, matching /add-dir behavior Added a /doctor check that proposes trimming checked-in CLAUDE.md files by cutting content Claude could derive from the codebase /commit-push-pr now auto-allows git push to the repo's configured push remote (remote.pu

v2.1.205

07/09 05:22

What's changed Added an auto mode rule that blocks tampering with session transcript files Fixed --json-schema silently producing unstructured output when the schema was invalid, and schemas using the format keyword being rejected Fixed a message sent while Claude was working being silently lost whe