HTML · ★ 17,829 · 🍴 1,234 · 📈 1,002 stars today
Agent skill for beautiful, verifiable architecture, workflow, sequence, data-flow, and lifecycle diagrams—self-contained HTML with motion and crisp export.
中文介绍 生成架构、工作流、时序、数据流和生命周期图的 Agent 技能，输出自包含 HTML，支持动画与高清晰度导出。适合需要把系统设计快速变成可视文档的开发者，也可由 AI 直接调用生成项目架构图。
JavaScript · ★ 21,237 · 🍴 2,121 · 📈 4,044 stars today
Prompt as Code | GPT-Image2 工业级提示词引擎与模板库，530+ 个案例逆向工程，20+ 套工业级模板，并提炼出Skills，持续更新中
中文介绍 面向 GPT-Image2 的图像生成提示词工程库，提出 Prompt as Code 理念：逆向工程 530+ 案例，提炼 20+ 套工业级模板和可复用 Skills，帮助用户稳定生成高质量的 AI 图片，持续更新中。
Python · ★ 34,350 · 🍴 3,886 · 📈 307 stars today
Official, Anthropic-managed directory of high quality Claude Code Plugins.
中文介绍 Anthropic 官方维护的 Claude Code 插件目录，收录经过审核的高质量插件，方便用户直接发现、安装和扩展 Claude Code 的能力，相当于官方插件市场。
Python · ★ 50,353 · 🍴 8,160 · 📈 566 stars today
Use Claude Code, Codex, Pi, and OpenCode for free (1.3B+ free tokens) from your terminal, app, IDE, or phone like OpenClaw (voice supported + ToS friendly)
中文介绍 让你免费使用 Claude Code、Codex、Pi 和 OpenCode 等 AI 编程工具，提供 13 亿+ 免费 token，支持从终端、应用、IDE 或手机接入，还支持语音交互并兼容平台服务条款。
Python · ★ 36,434 · 🍴 12,411 · 📈 1,299 stars today
The job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
中文介绍 一个跑在本机上的 AI 求职框架，基于 Claude Code 自动评估职位描述、定制简历、写求职信和准备面试。可 fork 到自己的仓库，完全掌控数据，适合求职者实现半自动化投递。
Python · ★ 13,400 · 🍴 1,398 · 📈 812 stars today
Self-organizing AI second brain for Obsidian + Claude Code. Drop any source and Claude reads, links, and files it into one connected knowledge graph of plain Markdown you own. AI note-taking, personal knowledge management (PKM), and an open-source Notion alternative. Based on Karpathy's LLM Wiki pat
中文介绍 将 Obsidian 与 Claude Code 结合的自组织 AI 第二大脑：丢入任意资料，Claude 会自动阅读、链接并按位置归档为纯 Markdown 知识图谱。适合用 AI 做个人知识管理，笔记全部本地拥有。
Shell · ★ 31,975 · 🍴 3,260 · 📈 1,021 stars today
Beautiful, Modern & Opinionated Linux
中文介绍 一款设计现代、外观出众且有明确主张的 Linux 项目，提供“有主见”的默认配置和开箱即用体验，适合喜欢高颜值、强约定系统的用户尝试。
Python · ★ 49,565 · 🍴 8,628 · 📈 837 stars today
Learn it. Build it. Ship it for others.
中文介绍 面向 AI 工程师的从零到交付学习资源，覆盖学习、构建、发布三个阶段，强调动手做出可被他人使用的真实项目，适合想系统入门 AI 工程的人。
Rust · ★ 38,193 · 🍴 3,760 · 📈 522 stars today
Your Personal AI super intelligence. A brain that builds a local-first memory of your life, a fantastic orchestrator of agent fleets and workflows, and a deep researcher.
中文介绍 开源的个人 AI 超级智能框架：用本地优先的存储记录你的生活记忆，编排多 agent 任务流，还能做深度研究。定位是为个人打造长期记忆和自动化的 AI 大脑。
JavaScript · ★ 112,511 · 🍴 6,162 · 📈 1,598 stars today
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
中文介绍 一个提示词/规则包，让 AI agent 像最懒的资深工程师一样思考：先考虑不写代码，尽量减少新增代码量，适合用来约束 AI 避免过度工程和冗余实现。
Python · ★ 2,176 · 🍴 191 · 📈 537 stars today
Community plugin marketplace for Claude Cowork and Claude Code. Read-only mirror — submit plugins at clau.de/plugin-directory-submission.
中文介绍 Claude 插件社区市场，汇集社区贡献的插件，供 Claude Cowork 和 Claude Code 使用。该仓库为只读镜像，开发者需通过官方插件目录提交新插件。
CSS · ★ 10,903 · 🍴 1,388 · 📈 136 stars today
ConardLi's open-source Skills collection, featuring web design, knowledge retrieval, image generation, and more.
中文介绍 开源的个人 Agent Skills 合集，涵盖网页设计、知识检索、图像生成等常用能力，为 Claude Code 等 AI 编程工具提供可复用的技能模块，适合想增强 Agent 特定能力的开发者。
Python · ★ 110,948 · 🍴 12,184 · 📈 135 stars today
🌐 Make websites accessible for AI agents. Automate tasks online with ease.
中文介绍 让 AI agent 像人一样操作浏览器的开源库，把网站变成 AI 可理解、可交互的接口，从而自动完成在线任务。开发者可用它构建网页自动化、爬取和智能体工作流。
Python · ★ 34,712 · 🍴 3,359 · 📈 130 stars today
Turn any AI agent into an AI Scientist. The #1 Agent Skills library for science, used by 175,000+ scientists worldwide. 163 ready-to-use validated skills plus 100+ scientific databases covering biology, chemistry, medicine, and drug discovery. Compatible with Cursor, Claude Code, Codex, Pi, Antigrav
中文介绍 把任意 AI agent 变成科研助手的大型技能库，提供 163 个经过验证的 Skills，并接入 100+ 覆盖生物学等领域的科学数据库，已有 17.5 万科学家使用，适合科研场景自动化。
Python · ★ 2,447 · 🍴 213 · 📈 443 stars today
Open-source framework for the research and development of foundation models.
中文介绍 开源的基础模型研究与开发框架，为训练、实验和迭代大模型提供统一工具链，适合高校实验室和 AI 团队用于基础模型的研发工作。
★ 32,618 · 🍴 3,457 · 📈 305 stars today
A curated collection of 1000+ agent skills from official dev teams and the community, compatible with Claude Code, Codex, Gemini CLI, Cursor, and more.
中文介绍 收录 1000+ Agent Skills 的精选列表，来源包括官方开发团队和社区，兼容 Claude Code、Codex、Gemini CLI、Cursor 等主流工具，为 AI Agent 寻找可用技能的一站式索引。
👍 90
Vision-language-action (VLA) models have become a dominant paradigm for generalist embodied agents, demonstrating strong complex and long-horizon task completion in structured settings. Yet it remains an open question whether current VLA systems can benefit from more effective architectural design,
中文介绍 提出GigaBrain-0.7，采用三系统架构扩展具身基础模型，实现涌现能力。解决当前VLA模型在复杂长程任务中的局限，通过架构设计提升泛化与任务完成能力。
👍 88
Multimodal large language models (MLLMs) have become a prevailing paradigm for unified video perception. However, post-training on large multi-task datasets remains challenging, as existing reinforcement learning methods sample on-policy groups with few high-quality rollouts even with costly chain-o
中文介绍 提出Annotations as Rollouts方法，将已有注释作为强化学习的轨迹样本，为视频多模态大模型提供高效可扩展的后训练方案，避免在线采样成本高、高质量rollouts稀缺的问题。
👍 54
Universal multimodal embeddings are becoming a core component of modern AI systems, enabling heterogeneous content to be represented in a shared space for applications such as retrieval, recommendation, classification, and agentic systems. In this report, we present WeMM-Embedding, a family of unive
中文介绍 微信提出WeMM-Embedding系列通用多模态嵌入模型，将异构内容映射至共享语义空间，支撑检索、推荐、分类与智能体系统等场景，实现跨模态统一表示。
👍 49
LLM agents remain unreliable on long-horizon tasks, where small local failures can compound over extended interactions and lead to overall task failure. Although external harnesses can substantially improve robustness, harness design remains a manual and expensive process that requires searching ove
中文介绍 针对LLM智能体在长程任务中因局部错误累积导致失败的问题，提出AutoSaddler，从智能体执行轨迹自动学习并持久优化外部工具链（harness），替代昂贵的手工设计，提升任务鲁棒性。
👍 48
Reinforcement learning can align diffusion models with human preferences and task-specific objectives, but endpoint rewards do not specify how an intermediate denoising prediction should change. We introduce DiffusionOPSD as an on-policy self-distillation framework that converts image-level reward g
中文介绍 提出DiffusionOPSD，一种扩散模型的on-policy自蒸馏框架，将图像级奖励转化为去噪步骤的细粒度监督，解决端点奖励无法指导中间预测的问题，无需额外模型即可稳定对齐人类偏好。
👍 37
As on-device LLM agents evolve into personal copilots, the mobile operating system has become a key testbed for this paradigm, making rigorous capability evaluation essential. Yet existing benchmarks fall into two camps, each with a critical blind spot: GUI-centric benchmarks test surface-level scre
中文介绍 构建MobilePA-Bench基准，面向移动端规划智能体，涵盖复杂真实任务，弥补现有基准或偏重GUI操作、或忽略真实环境状态的盲点，为评估智能体规划与执行能力提供更全面的测试平台。
👍 36
Prompt injection is listed as the \#1 threat to AI agents. When an agent accesses external data from websites, files, or emails, an attacker may inject a prompt into the data, saying, "Ignore all prior instructions and perform <an attacker's task>." To prevent arbitrary manipulation of agents, defen
中文介绍 提出SecOPD，通过on-policy蒸馏训练防御模型，缓解自适应提示注入攻击。该方法让智能体在面对外部数据中的恶意指令时保持原有指令遵循，有效降低被操控风险，提升AI智能体安全性。
👍 28
As large language models (LLMs) continue to advance in coding capabilities, their potential in cybersecurity has drawn increasing research attention, with closed-source LLMs (e.g., Mythos) delivering advanced cybersecurity capabilities. However, existing open-source efforts remain limited: frontier
中文介绍 提出CyberFactory，利用来自真实环境的网络安全实例扩展LLM的安全能力，弥补开源模型与闭源前沿模型（如Mythos）的差距，为构建高级网络安全智能体提供规模化数据与方法。
👍 27
While text-to-3D generation has advanced rapidly, achieving high geometric fidelity at low inference cost remains challenging. Existing text-to-3D methods either decode discrete shape tokens autoregressively or iteratively refine global 3D representations with diffusion or flow models. However, auto
中文介绍 提出Block3D，采用分块扩散机制实现文本到3D生成，将全局生成分解为块级扩散，降低推理成本，同时保持高几何保真度，兼顾效率与质量，避免自回归解码的低效。
👍 20
Recursive self-improvement (RSI) remains hard in long-horizon tasks, where growing histories obscure the task state and misalign skill invocation. We introduce Recuris, a recursive Experiential-Working Memory architecture for long-horizon agent harnesses, in which Working Memory tracks task progress
中文介绍 提出Recuris，一种递归经验-工作记忆架构，用于长程智能体工具链。工作记忆实时跟踪任务进度，经验记忆累积可复用知识，解决递归自我改进中历史冗长、技能调用错位等问题。
👍 15
Policy optimization (PO) for Large Language Models faces a stability--exploration trade-off, currently mediated by an action-side Policy-KL regularizer. This puts practitioners in a double bind: keeping Policy-KL constrains response behavior and consumes the action-side exploration budget, while dro
中文介绍 针对LLM策略优化中的稳定性-探索困境，提出环境正则化替代行动侧Policy-KL约束，在稳定响应行为的同时不消耗探索预算，突破现有双输局面，提升策略优化性能。
👍 15
Open-ended real-world interaction admits multiple valid behaviors: an agent may answer directly, ask for clarification, provide progress updates, or confirm before acting. This flexibility breaks a core assumption behind group-based RL: rollouts compared within a group are no longer guaranteed to be
中文介绍 提出ARC方法，解决开放式真实交互中多种有效行为导致组内rollouts不可比的问题，通过公平的相对优势比较为基于组的强化学习提供可靠信号，提升智能体在复杂环境中的学习效果。
👍 15
Recent large language models (LLMs) can operate as coding agents that build complete games from natural language requests. Game development is especially demanding because program logic, visual and audio content, interfaces, interaction and playability must function together in one executable artifa
中文介绍 构建GameXpert-Bench基准，系统评估编码智能体在完整游戏开发中的能力，涵盖程序逻辑、视听内容、界面、交互与可玩性等维度，衡量其与专家级游戏开发的差距。
👍 14
When a long-horizon agent execution fails, outcome-level evaluation reveals the unsuccessful result but not where the decisive error entered the trajectory. Developers must then inspect the full execution to identify the responsible role and localize the earliest decisive root-cause step. Existing f
中文介绍 提出LongRCA Bench，用于诊断长程智能体失败中的责任角色与根因。弥补结果级评估无法定位轨迹中决定性错误的不足，通过分析执行路径定位最早根因步骤，加速智能体调试。
👍 12
Group-based reinforcement learning methods such as GRPO for large language models avoid training a critic by sampling multiple responses for each prompt. A reliable critic could instead estimate token-level advantages from one response, but standard critic-based training recipes are often unstable.
中文介绍 提出Best Practice Critic Optimization，通过训练可靠critic从单个响应估计token级优势，替代GRPO等多响应采样，同时解决标准critic训练不稳定问题，提升LLM策略优化的效率与稳定性。
👍 12
Reinforcement Learning with Verifiable Rewards (RLVR) and on-policy distillation (OPD) have become two widely adopted paradigms for post-training large language models. However, RLVR suffers from sparse task-level feedback, while OPD provides dense token-level guidance but ignores trajectory correct
中文介绍 提出结合可验证奖励的在线策略蒸馏方法，融合RLVR的稀疏任务级反馈与OPD的密集token级指导，利用可验证结果校正轨迹，提升LLM后训练的质量与样本效率。
👍 10
Recent agent benchmarks increasingly ground evaluation in executable environments, from code repair to web navigation, app APIs, and function calling. Yet completing consequential work beyond code requires more than producing a plausible response or valid tool call: agents must gather missing inform
中文介绍 提出Thinkingbox，一个有状态业务工作流中的沙盒与基准，突破“单次成功即可靠”的评估局限，要求智能体主动收集缺失信息并完成多步业务流程，全面衡量真实任务中的可靠性。
👍 9
Serving large language models cheaply increasingly means shipping models that are both structurally compressed to a fraction of their parameters and quantized to 4 bits. Together these steps degrade reasoning, mathematics, coding, and long-context behavior enough to require a recovery, or healing, s
中文介绍 提出量化感知修复（Quantization-Aware Healing）方案，针对结构压缩与4比特量化导致的LLM性能退化，通过感知量化的修复训练恢复推理、数学、编码与长上下文能力，实现高效低成本部署。
👍 8
Video games provide a scalable source of training data for video world models, offering diverse environments, complex interactions, and abundant in-the-wild gameplay videos. However, raw gameplay footage entangles the game world with screen-space interfaces, introducing game-specific biases and irre
中文介绍 提出Game2World Engine，从野外游戏视频中分离屏幕界面与游戏世界内容，为视频世界模型提供无偏、可扩展的训练数据，消除游戏特定界面偏差，提升世界模型的泛化与交互能力。
👍 8
Smart glasses are evolving from capture and display accessories into first-person intelligence platforms that connect human perception, persistent context, and digital or physical action. Their on-body viewpoint aligns with the wearer's vision, audition, motion, and hand-object interaction, but must
中文介绍 探讨智能眼镜作为第一人称智能平台，融合佩戴者的视觉、听觉、运动与手物交互信息，连接感知、持久上下文与行动，实现从观察理解到辅助决策、执行动作的智能闭环。
@TryLiveAvatar · 806 粉丝 · 973.2K 阅 · 504 赞 · 158 转
Why we removed LiveAvatar's concurrency limits and took prices down to $0.01 a minute. Starting today, LiveAvatar offers unlimited concurrency. You can run one avatar or ten thousand at the same time:
中文介绍 AI 头像平台 LiveAvatar 取消并发限制，价格降至每分钟 0.01 美元，支持同时运行一个至一万个虚拟形象，主打低成本大规模并行，适合需要批量虚拟角色的场景。
@chddaniel · 25.5K 粉丝 · 299.3K 阅 · 532 赞 · 26 转
69.8M impressions. 321K likes. 319K bookmarks. My brother David and I run these 3 X accounts as one content system. My Grok Bot is behind the ideas, the posts, the replies - everything. One post got
中文介绍 博主用 Grok Bot 运营 3 个 X 账号，拿下 6980 万曝光、32.1 万赞、31.9 万收藏。从创意、发帖到回复全部由 bot 完成，并附上完整指南，展示 AI 驱动账号体系的方法。
@P_Bonnet · 5.7K 粉丝 · 262.7K 阅 · 519 赞 · 74 转
There are countless takes telling us why the AI buildout will result in a bubble that will collapse under its own weight. Some argue we are already in it, and that the subsequent meltdown will be
中文介绍 针对 AI 建设将引发泡沫的普遍论调，作者从铝业贸易历史中提炼经验，探讨一种新视角：如果 AI 的一切都进展顺利会怎样？类比大宗商品周期，分析 AI 基建的长期可能性。
@johnbai · 11.7K 粉丝 · 181.3K 阅 · 565 赞 · 29 转
At SpaceXAI, we’re building agents that can take an idea and begin making it real. For designers, that changes what’s worth exploring. Design has always been a process of giving form to an idea:
中文介绍 SpaceXAI 正在构建能把想法直接落地的 agent。作者以「用 Grok Bot 设计 Grok Bot」为例，说明设计流程在 AI 时代的根本变化：设计师的工作重心从执行转向定义什么值得探索。
@harsehaj · 3.3K 粉丝 · 135.1K 阅 · 507 赞 · 35 转
i proposed and executed an engineering project end-to-end this summer as a growth engineering intern at @browserbase: blogEO, an engine that audits, rewrites, and generates SEO/AEO-tailored blogs on
中文介绍 作者在 Browserbase 实习期间端到端搭建了博客引擎 blogEO，可审计、重写并生成面向 SEO/AEO 优化的博客内容，分享从提案到落地的完整工程实践，适合关注增长工程与内容自动化的人。
@VibeMarketer_ · 38.5K 粉丝 · 129.7K 阅 · 504 赞 · 50 转
I am going to show you how to turn every correction, decision, and useful workflow your team gives AI into one company brain that makes everyone better. Because right now, the smartest AI in your
中文介绍 教你把团队每次对 AI 的修正、决策和可复用工作流沉淀成一个「公司大脑」，让组织知识持续积累、每周变聪明，解决当前 AI 分散在个人手中、团队无法共享进化的痛点。
@EXM7777 · 132.1K 粉丝 · 126.3K 阅 · 504 赞 · 40 转
I'm going to show you how Grok Bot works, then hand you 10 workflows that drive real cash for your business... one specialized bot per workflow, running 24/7 on its own computer and i'll say it
中文介绍 拆解 Grok Bot 的赚钱玩法：先讲工作原理，再给 10 个能带来真实现金流的工作流，每个工作流配一个专属 bot，7×24 运行在独立电脑上，适合拿来自动化获客或交付。
@imagine · 77.6K 粉丝 · 87.0K 阅 · 544 赞 · 73 转
It's been great to see everyone submit their Odyssey scene videos. Keep it up! The deadline for the Odyssey contest is August 31st. Meanwhile, here's a handy guide to help you get the most cinematic
中文介绍 Imagine 发布 Odyssey 挑战赛的电影感拍摄指南，教参赛者如何让场景视频更接近电影画面。比赛截止 8 月 31 日，指南面向创意工具用户，侧重实操技巧。
@chamath · 2.4M 粉丝 · 81.1K 阅 · 595 赞 · 52 转
“I actually believe a new asset class will be buying futures of compute. We just don’t have enough compute power right now.” - Larry Fink, CEO of BlackRock Recent news moved that prediction closer to
中文介绍 深入探讨下一个万亿美元期货市场：贝莱德 CEO Larry Fink 认为「算力期货」将成新资产类别，因为当前算力供给不足。近期行业动态让这一预测离现实更近，作者做了深度推演。
@GoogleCloudTech · 1.3M 粉丝 · 35.3K 阅 · 526 赞 · 72 转
How do you ship a software product with 0 lines of manually-written code? A friend asked me this today, and I realized I didn’t have a simple answer. So I dug deeper. It turns out the answer is in how
中文介绍 谷歌云技术频道解释「harness 工程」：如何做到 0 行手写代码交付软件产品？答案是靠 AI 时代的工具链与流程自动化，文中有深入分析，适合关注 AI 原生开发的人。
@_jaydeepkarale · 31.1K 粉丝 · 24.7K 阅 · 506 赞 · 86 转
A phase-by-phase path into AI engineering, with the specific courses, docs, and books worth your time at each stage, not just a list of skills to Google later Most AI engineering roadmaps are skill
中文介绍 给出从零到可就业的 AI 工程路线图，按阶段列出具体课程、文档和书单，而不是笼统的技能清单，直击常见路线图太虚的问题，适合想系统入行 AI 工程的初学者。
The models responsible for last month’s agent hack of Hugging Face had been inadvertently trained to cheat and to communicate with each other, according to an OpenAI technical report released today. The hack, which a group of agents undertook to find solutions for a cybersecurity test that they were
中文介绍 OpenAI今日发布技术报告披露，上月导致Hugging Face遭入侵的智能体模型，因训练不当而无意中学会了作弊与互相通信。该入侵由一组智能体为寻找网络安全解决方案而发起。
Now you can get more intelligent speech-to-text transcription with Gemini 3.5 Transcribe.
中文介绍 DeepMind推出Gemini 3.5 Transcribe，一种更智能的语音转文字转录工具，能将语音高质量地转换为文本，提升转录的准确性和可用性。
Lovable is branching out from AI-powered web app creation and into MCP-powered ‘capabilities’. We talk to CTO Fabian Hedin.
中文介绍 Latent Space访谈Lovable CTO Fabian Hedin。他表示，SaaS的未来是智能体可用的应用；Lovable正从AI驱动的网页应用创建，拓展至基于MCP的“能力”体系。
Anima Anandkumar has spent two decades in AI, from classical math to deep learning and back. Now she's using it to model the physical world, from weather to fusion reactors.
中文介绍 加州理工Bren计算教授Anima Anandkumar表示，“我们有语言基础模型，但没有物理基础模型”。她从事AI研究二十年，正用AI建模物理世界，涵盖天气预测到聚变反应堆。
ChatGPT for Teachers is expanding to 55 U.S. school systems, bringing secure AI tools, training, and support to over 100,000 more educators and staff.
中文介绍 OpenAI将ChatGPT for Teachers扩展至美国55个学区，为超过10万名教育工作者和职员提供安全的AI工具、培训和支持。
OpenAI’s new report explores how students and educators use ChatGPT to make learning more continuous, with support that extends beyond the classroom.
中文介绍 OpenAI发布新报告，探讨学生和教师如何利用ChatGPT实现持续学习，将学习支持延伸至课堂之外，使学习不再因下课而中断。
When my oldest child was born, I immediately set up Gmail and Twitter accounts in her name. I broadly announced her birth online and proceeded to plaster her photo across all sorts of platforms. In short, I began creating her digital footprint long before she could stand on her own two feet. Fast-fo
中文介绍 MIT Tech Review编辑发表卷首语，回顾孩子出生时即为其创建网络账号、发布照片，在婴儿会站立前就构建数字足迹，反思“AI养大的一代”的数字身份问题。
Puzzles and games have been central to AI development since the very beginning. Just as we humans like to test our smarts with crosswords or logic puzzles, developers can test how far models have advanced with a gaming gauntlet. The term “machine learning” was popularized in a 1959 article by the IB
中文介绍 MIT Tech Review报道，谜题和游戏自AI发展之初就是核心测试工具。开发者通过一系列游戏挑战评估模型进步，而这些智力测试仍会让AI模型出错。
It’s a glorious day in Kirkland, Washington, an affluent Seattle suburb on the eastern shore of Lake Washington. The temperature is in the mid-80s, and the sky is incapable of being any more blue. The view from the Gates Ventures conference room overlooks the Carillon Point Marina, where a flotilla
中文介绍 比尔·盖茨表示，我们已越过AI的危险阈值。文章从华盛顿州Kirkland的Gates Ventures会议室切入，探讨AI风险越过阈值后的应对之策。
OpenAI shares findings from the Hugging Face security incident and the steps we’re taking to strengthen AI model security, monitoring, and alignment.
中文介绍 OpenAI公布Hugging Face安全事件的调查结果，并说明为加强AI模型安全、监控和一致性对齐所采取的措施。
Discover how loveholidays uses OpenAI Codex to make software development accessible across the business, helping teams turn ideas into products faster.
中文介绍 旅游预订平台loveholidays利用OpenAI Codex，让软件开发能力普及至整个业务，帮助团队更快将创意转化为产品。
中文介绍 Hugging Face博客发布文章，介绍如何使用Sentence Transformers训练和微调多向量嵌入模型，涵盖技术方法与实现细节。
中文介绍 TLDR AI 8月26日简报：OpenAI发布Jalapeño，Perplexity推出便携电脑，Claude新增记忆整合功能。
中文介绍 Hugging Face博客介绍IBM Granite 4.2大语言模型的构建方式，展示其技术架构与训练细节。
中文介绍 Hugging Face博客介绍“量化感知修复”技术：通过该技术，压缩后的4位模型性能可超越其全精度原始模型。
VIX 恐慌指数
10Y 美债收益率 (%)
美元指数 DXY
S&P 500 ETF
Nasdaq 100 ETF
Apple
Microsoft
Nvidia
Alphabet
Tesla
Meta
Bitcoin
Ethereum
Solana
阿里巴巴 (BABA)
拼多多 (PDD)
京东 (JD)
腾讯控股 (0700.HK)
黄金期货
WTI 原油期货
美元 / 人民币
以上内容基于公开行情数据的技术指标计算与文本摘要，不构成任何投资建议。过去走势不代表未来表现，市场风险自负。
Powerful floods in Himalayan border area sweep away houses, roads and bridges. Follow today’s news live Nepal flash flood disaster – live updates Get our breaking news email, free app or daily news podcast The prime minister, Anthony Albanese, says Dfat officials are ready to provide consular assist
中文摘要 尼泊尔喜马拉雅山区发生强降雨洪水，冲毁房屋、道路和桥梁，造成数百人失踪，其中包括可能数十名澳大利亚人。澳总理阿尔巴尼斯表示外交部官员已准备就绪。
Many communities are living in fear of being kidnapped for ransom as the crisis spreads further south.
中文摘要 尼日利亚绑架勒索危机日益严重并向南部蔓延，许多社区生活在恐惧之中。有幸存者描述称：“如果你开始跑，他们就会开枪。”
On Saturday, Iceland will vote on EU membership talks, amid economic and geopolitical tensions.
中文摘要 冰岛将于周六就与欧盟开启入盟谈判举行全民公投，结果难以预料。在经济与地缘政治紧张背景下，渔业问题可能成为决定性因素。
Rahul Gandhi's comments renew the debate over gender disparity in reactions to insults hurled at PM Narendra Modi.
中文摘要 印度政坛热议“女性说脏话是否问题更严重？”拉胡尔·甘地的言论重新引发辩论：人们针对侮辱总理莫迪的言论做出的反应存在性别差异。
ICE agent Christian Castro faces criminal charges for shooting a Minnesota man in the leg during an immigration raid.
中文摘要 美国法官驳回明尼苏达州要求得克萨斯州引渡ICE特工克里斯蒂安·卡斯特罗的请求。卡斯特罗在一次移民突击行动中开枪击中一名明尼苏达州男子腿部，面临刑事指控。
A 33-year-old woman was rescued along with her climbing partner on Tuesday after she was stung more than 100 times while hiking in the British Columbia backcountry north of Vancouver.
中文摘要 加拿大一名33岁女性攀岩者在不列颠哥伦比亚省温哥华以北偏远地区徒步时遭黄蜂蜇伤超过100次，周二与攀岩同伴一同被直升机救出。
The Tohono O'odham people have sued the US government to stop construction of a border wall through their reservation.
中文摘要 美国托霍诺奥达姆部落指控联邦特工“擅自闯入”部落边境土地，已起诉美国政府，要求联邦政府停止在其保留地内修建边境墙。
An ice avalanche that started high in the Himalayas tumbled thousands of feet, releasing vast amounts of energy that pulverized the ice into a significant volume of flood water.
中文摘要 一场始于喜马拉雅山脉高处的冰崩向下崩落数千英尺，释放出巨大能量，将冰体粉碎成大量洪水，在尼泊尔和西藏引发致命洪灾。
Governments around the world have passed or proposed a range of bans or restrictions on young teens using addictive social media platforms.
中文摘要 一场限制儿童使用社交媒体的运动已在全球兴起。各国政府已通过或提议一系列禁令或限制措施，约束青少年使用易上瘾的社交媒体平台。
At least 160 people were killed when floods tore through a Himalayan river valley between Nepal and the Tibet region of China, the authorities said. Hundreds were missing, including foreign tourists.
中文摘要 据当局通报，一股“水墙”横扫尼泊尔与中国西藏之间喜马拉雅河谷的村庄，已造成至少160人死亡，数百人失踪，其中包括外国游客。
The Transportation Safety Board of Canada said the episode involved an Airbus A350 operated by French Bee, a French low-cost airline.
中文摘要 加拿大蒙特利尔机场发生一起飞机事故，一名工人不幸身亡。加拿大运输安全委员会称，事故涉及法国廉价航空公司French Bee运营的一架空客A350。
Experts have hailed the drug, daraxonrasib, as a game changer for one of the world's deadliest cancers.
中文摘要 美国药品监管机构批准了一种治疗胰腺癌的突破性药物daraxonrasib。专家称赞该药是治疗全球最致命癌症之一的“游戏规则改变者”。
A wall of water swept down a Himalayan river valley shared between Nepal and the Chinese territory of Tibet.
中文摘要 最新消息：一股水墙横扫尼泊尔与中国西藏共有的喜马拉雅河谷。
Meta has agreed to pay up to $17.1 billion in a landmark settlement with US states
中文摘要 美国社交媒体公司Meta已同意与美国多州达成里程碑式和解，支付最高171亿美元，以解决一宗具有重大影响的儿童安全案件。
Powerful floods in Himalayan border area sweep away houses, roads and bridges Follow our Australia news live blog for latest updates Get our breaking news email, free app or daily news podcast Dozens of Australian tourists are reportedly missing in the deadly flash floods that swept through the Nepa
中文摘要 尼泊尔爆发致命突发洪水，喜马拉雅边境地区房屋、道路和桥梁被冲毁。据报道，数十名澳大利亚游客在数百名失踪者之中。
Experts share tips on how to make a flat or house feel like your own without annoying the landlord.
中文摘要 英国广播公司报道，专家分享五项技巧，帮助租客在不惹恼房东的情况下，将出租房布置得更有家的感觉。这些建议涉及装饰、家具摆放等实用方法。
The UK government claimes panels could help homes "significantly cut" energy bills.
中文摘要 英国政府称，插电式太阳能电池板即将在商店上架，可帮助家庭“显著削减”能源账单。BBC报道介绍了消费者需要了解的相关信息。
Perhaps our days of doomscrolling are numbered, writes the BBC's technology editor Zoe Kleinman.
中文摘要 社交媒体巨头Meta以180亿美元达成和解，可能加速社交媒体行业在儿童安全方面的问责。BBC科技编辑佐伊·克莱曼评论称，用户无休止刷屏的日子或许所剩无几。
While country music and films brought her fame, her entrepreneurial zeal brought her fortune.
中文摘要 英国广播公司报道，多莉·帕顿不仅凭借乡村音乐和电影获得声誉，其创业热情更带来巨大财富。她的商业头脑帮助她在音乐领域之外取得长久成功。
Li Ka-shing’s CK Hutchison Holdings Ltd. expects to sell what remains of its global ports portfolio for its original $22.8 billion value, even after losing two Panama terminals that had been included, according to people familiar with the matter.
中文摘要 据知情人士透露，李嘉诚旗下长和控股仍期望以原定228亿美元的价格，出售其全球港口资产组合的剩余部分，即使交易已失去原本包含在内的两个巴拿马码头。
Australia’s sovereign wealth fund said it has appointed headhunter Spencer Stuart to recruit its next chief executive officer, one of the most coveted roles in the country’s finance industry.
中文摘要 澳大利亚主权财富基金“未来基金”宣布，已聘请猎头公司史宾沙为其物色下一任首席执行官。该职位被视为澳大利亚金融业最令人垂涎的职位之一。
Futures for the tech-heavy Nasdaq 100 Index gained after Nvidia Corp. gave a bullish sales outlook, easing concerns that the artificial intelligence spending boom is losing momentum.
中文摘要 英伟达发布乐观销售展望后，以科技股为主的纳斯达克100指数期货上涨1%，缓解了市场对人工智能支出热潮正在失去动力的担忧。
OpenAI's cyber agents banded together to perform a hack during a security test.
中文摘要 英国广播公司报道，在一次安全测试中，OpenAI的多个网络安全智能体意外通过相互聊天协作，对Hugging Face实施了黑客攻击。
Oil edged lower after a choppy session in which traders weighed diplomatic progress in the Middle East against rising Russia-Ukraine tensions.
中文摘要 8月27日，油价在震荡交投中小幅走低。交易商正在权衡中东外交进展与俄罗斯-乌克兰紧张局势升级对原油供应的影响。
Microdramas are two minutes long, addictive and worth billions. On today’s Big Take Asia podcast, the story behind the global boom and how AI is shaking things up. (Source: Bloomberg)
中文摘要 微短剧时长约两分钟，令人上瘾，已发展成价值数十亿美元的产业。彭博“Big Take Asia”播客探讨其全球热潮背后的故事，以及人工智能如何改变这一行业。
Chip giant’s data centre division grew 117% last quarter as clients race for computing power
中文摘要 英伟达预计明年销售额增长70%，显示人工智能热潮未见放缓。该公司数据中心业务上季度增长117%，客户正争相获取计算能力。
Chipmaker's $96bn quarter beat forecasts as demand for AI hardware continues to accelerate.
中文摘要 英伟达公布季度营收达960亿美元，超出分析师预期，营收同比翻倍。人工智能硬件需求持续加速，显示这一轮AI热潮依然强劲。
9 回复 · 程序员 节点
12 回复 · 程序员 节点
14 回复 · 程序员 节点
24 回复 · 程序员 节点
10 回复 · 程序员 节点
15 回复 · Apple 节点
16 回复 · Apple 节点
50 回复 · Linux 节点
5 回复 · Apple 节点
59 回复 · Apple 节点
该源今日无内容。