JavaScript · ★ 153,401 · 🍴 8,226 · 📈 1,281 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 让 AI 编程助手像团队里最懒的资深工程师那样思考：能复用就不重写，能不写的代码就不写。它通过约束 agent 的生成倾向，减少冗余实现与过度设计，适合想控制 AI 代码膨胀、追求最小改动的开发者日常使用。
JavaScript · ★ 75,295 · 🍴 4,516 · 📈 699 stars today
The design language that makes your AI harness better at design.
中文介绍 面向 AI 编程工具的设计语言与规范，让 Claude Code 等 agent 在生成界面时更懂排版、间距与视觉层级。适合用 AI 快速搭前端、却苦于产出界面千篇一律的开发者，可作为 harness 的设计层配置使用。
JavaScript · ★ 272,243 · 🍴 40,652 · 📈 897 stars today
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
中文介绍 针对 AI agent 运行框架的性能优化系统，覆盖技能、直觉、记忆、安全与调研优先的开发流程，可接入 Claude Code、Codex、Opencode、Cursor 等工具。适合重度使用多个编程 agent、需要统一管理与加速的工程师。
TypeScript · ★ 16,814 · 🍴 813 · 📈 302 stars today
Build production-ready applications in TypeScript
中文介绍 面向 TypeScript 的生产级应用开发框架，用类型化的 Effect 系统统一处理异步、错误、依赖注入与并发，把副作用收进可组合的抽象中。适合构建大型后端服务、对可靠性与可测试性要求高的 TS 团队。
Go · ★ 109,533 · 🍴 6,334 · 📈 507 stars today
🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.
中文介绍 用“原始人说话”的极简提示风格压缩 AI 编程 agent 的 token 消耗，号称可省约 65%。提供 skill 与 proxy 两种接入方式，适合长会话、高频调用大模型的开发者降低成本与延迟。
Python · ★ 89,799 · 🍴 7,902 · 📈 1,696 stars today
Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
中文介绍 给 AI agent 装上“眼睛”，通过单个 CLI 读取并搜索 Twitter、Reddit、YouTube、GitHub、Bilibili、小红书等平台内容，无需逐个申请 API、零费用。适合需要实时外部信息做调研与内容分析的 agent 工作流。
TypeScript · ★ 24,684 · 🍴 6,444 · 📈 252 stars today
中文介绍 由 T3 Stack 生态作者 pingdotgg 发布，仓库未附带说明文字，从命名看与 T3 技术栈的代码生成或编码工具有关。具体能力需查阅仓库源码与 README，适合关注 T3 生态动态的开发者跟进。
TypeScript · ★ 95,572 · 🍴 8,459 · 📈 79 stars today
Persistent Context Across Sessions for Every Agent – Captures everything your agent does during sessions, compresses it with AI, and injects relevant context back into future sessions. Works with Claude Code, OpenClaw, Codex, Gemini, Hermes, Copilot, OpenCode + More
中文介绍 为各类 AI agent 提供跨会话的持久记忆：记录会话中 agent 的全部操作，用 AI 压缩成摘要，并在后续会话中自动注入相关上下文。兼容 Claude Code 等工具，适合长期跑项目的开发者，免去反复交代背景。
TypeScript · ★ 10,565 · 🍴 1,282 · 📈 85 stars today
Agent workspace built on Cloudflare Workers for creating documents, building apps, and running agents with your company’s context and systems.
中文介绍 基于 Cloudflare Workers 构建的 agent 工作空间，可创建文档、搭建应用并运行 agent，同时接入企业自有的上下文与内部系统。适合已使用 Cloudflare 生态、想把 AI agent 落到公司内部流程的团队。
JavaScript · ★ 100,835 · 🍴 10,595 · 📈 252 stars today
Production-grade engineering skills for AI coding agents.
中文介绍 由 Addy Osmani 整理的面向 AI 编程 agent 的生产级工程技能集，把代码审查、测试、重构等成熟实践沉淀为可复用技能。适合希望 agent 产出更贴近真实工程标准的开发团队参考使用。
Shell · ★ 294,913 · 🍴 26,354 · 📈 577 stars today
An agentic skills framework & software development methodology that works.
中文介绍 一套 agent 技能框架与配套软件开发方法论，主张用可组合的技能定义 agent 行为，让 AI 参与编程的过程可复现、可协作。适合把 agent 纳入正式开发流程、而非零散临时提问的团队。
Shell · ★ 275,366 · 🍴 23,105 · 📈 751 stars today
Skills for Real Engineers. Straight from my .agents directory.
中文介绍 TypeScript 教育者 Matt Pocock 从其 .agents 目录直接开源的技能集合，面向真实工程场景，覆盖日常编码中的常见任务。适合想借鉴成熟开发者如何配置 AI agent 技能的人直接取用。
TypeScript · ★ 25,245 · 🍴 1,807 · 📈 256 stars today
Context window optimization for AI coding agents. Sandboxes tool output (98% reduction), persists session memory, and enforces routing across 17 platforms via MCP + hooks.
中文介绍 面向 AI 编程 agent 的上下文窗口优化工具：沙箱化工具输出可减少约 98% 的 token 占用，持久化会话记忆，并通过 MCP 与 hooks 在 17 个平台上统一路由。适合长任务、多平台 agent 用户控制上下文成本。
TypeScript · ★ 112,156 · 🍴 14,229 · 📈 408 stars today
AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI
中文介绍 AI agent 工具包，包含统一的 LLM API、agent 循环、终端 TUI 以及编码 agent 命令行工具。适合想自行搭建或定制编程 agent、不愿被单一厂商 SDK 绑定的开发者，作为底层框架使用。
Python · ★ 45,203 · 🍴 4,898 · 📈 214 stars today
Developer-first error tracking and performance monitoring
中文介绍 老牌开源错误追踪与性能监控平台，帮助开发者实时捕获异常、定位慢请求与前端性能问题，并提供 issue 归因与告警。适合各类 Web 与移动应用团队接入 CI/CD，做线上质量保障。
TypeScript · ★ 149,219 · 🍴 25,362 · 📈 128 stars today
Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflows - all through natural language commands.
中文介绍 Anthropic 官方推出的终端内 agentic 编程工具，能理解整个代码库，代为执行常规任务、解释复杂代码并处理 git 工作流。适合习惯命令行、希望把重复编码与版本控制操作交给 AI 的开发者。
Python · ★ 9,401 · 🍴 2,069 · 📈 193 stars today
中文介绍 jamwithai 出品的 Agentic RAG 实战课程仓库，聚焦如何把检索增强生成从演示做到生产可用。仓库暂无描述，从命名看包含课程代码与讲解材料，适合想系统学习 RAG 工程化落地的开发者。
Python · ★ 8,737 · 🍴 1,523 · 📈 44 stars today
中文介绍 美团 LongCat 团队开源的视频生成项目，仓库未提供描述。从命名判断属于其 LongCat 模型系列中的视频方向工作，适合关注视频生成进展、想在自有业务中部署或微调视频模型的研究者与开发者。
TypeScript · ★ 91,614 · 🍴 9,057 · 📈 232 stars today
The open-source CapCut alternative
中文介绍 开源版 CapCut（剪映）替代方案，提供视频剪辑所需的界面与基础能力，代码开放，可自行部署和二次开发。适合不愿受商业剪辑软件订阅与隐私条款限制的创作者，以及需要嵌入剪辑功能的团队。
👍 160
Streaming video LLMs must retain evidence before its relevance to future tasks is known and respond when sufficient evidence becomes available. The challenge is to form reusable factual memory without compromising real-time perception. We introduce OneStreamer, which jointly learns query-independent
中文介绍 提出 OneStreamer，统一流式视频交互中的感知、记忆与主动响应。针对流式视频 LLM 需在未来任务相关性未知时保留证据、并在证据充分时响应的难题，模型联合学习与查询无关的事实记忆构建，在不牺牲实时感知的前提下形成可复用记忆。
👍 153
On-policy learning has been argued to reduce catastrophic forgetting, produce sparser parameter updates, and improve generalisation. However, existing comparisons between supervised fine-tuning and reinforcement learning vary many factors simultaneously, making the contribution of rollout policy dif
中文介绍 系统研究 on-policy 与 off-policy 蒸馏的动态差异。现有 SFT 与 RL 对比同时改变多个因素，无法分离 rollout 策略本身的贡献；该工作控制变量拆解各因素，考察策略来源对灾难性遗忘、参数更新稀疏性与泛化能力的影响。
👍 57
Masked diffusion models (MDMs) generate sequences by progressively unmasking several tokens per denoising step, but their reverse process is typically factorized over positions, limiting sample quality in the few-step regime where diffusion's speed advantage over autoregressive decoding matters most
中文介绍 针对 masked diffusion 模型每步解掩多个 token、反向过程按位置因式分解，导致少步采样质量受限的问题，提出 E-MoE，将增强的 MoE 引入非因式分解扩散语言模型，在保留扩散相对自回归解码速度优势的同时提升生成质量。
👍 49
Multi-reward reinforcement learning trains large language models to satisfy multiple behavioral objectives simultaneously. Reward-wise normalization, as used in GDPO, preserves reward-specific relative information within rollout groups, but different objectives can still exhibit uneven learning prog
中文介绍 多奖励 RL 训练 LLM 同时满足多种行为目标，GDPO 式按奖励归一化虽保留组内相对信息，但不同目标学习进度不均、稀疏目标易被淹没。论文提出密度感知的奖励聚合方法，按奖励密度调整权重，改善稀疏奖励下的多目标优化。
👍 44
Real-world embodied tasks, from everyday activities to professional procedures, require agents to act under physical constraints while tracking evolving object and task states. Tool use sits at the heart of such tasks, as many everyday and professional activities are tool-mediated. Understanding the
中文介绍 工具使用是日常与专业活动的核心，但现有模型难以在第一人称视频中跟踪物体与任务状态。EgoTools 面向真实第一人称视频的工具中心推理，构建要求智能体在物理约束下操作工具、追踪状态演化的任务与评测。
👍 43
We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was never trained for by repurposing its existing skills, improving from its own attempts, and retaining what it learns, without retraining. Our key insight is that a broad controller already holds much of t
中文介绍 研究人形机器人 loco-manipulation 的测试时进化：无需重新训练，复用控制器已有技能，从自身尝试中改进并保留所学，完成从未训练过的任务。核心思路是在测试时进化奖励程序，通过迭代自我改进实现技能重组与新任务适应。
👍 41
Decentralized multi-agent path finding (MAPF) with communication requires agents to reach individual goals without collisions under partial observability. Learnable policies trained on expert data provide an effective approach to this problem. However, when several coordinated joint actions are vali
中文介绍 针对部分可观测、需通信的去中心化多智能体路径规划（MAPF），基于专家数据训练的策略在多个协调联合动作均可行时难以抉择。论文提出「主脑」式迭代意图去噪机制，逐步细化各智能体意图以联合优化动作，减少碰撞。
👍 38
Multi-step agents are trained on flat action streams: SFT and RLVR weight every token uniformly and ignore the sub-procedures that recur across tasks, the hierarchy that lets humans plan top-down from reusable routines. This structure sits unused, and flat training uses each scarce trajectory less f
中文介绍 多步 agent 通常在扁平动作序列上训练，SFT 与 RLVR 对所有 token 均匀加权，忽略了跨任务复现的子过程。X-Tree 将可复用经验 token 化为层级结构，显式建模「例程—子过程」层次，提升轨迹利用效率与 agent 泛化能力。
👍 34
Vision language model (VLM) agents can control robots through visual feedback and action primitives, but repeated model invocations and redundant observations incur substantial token overhead. We introduce PyRUA-Lean, an interactive code-execution framework that couples feedback-driven primitive com
中文介绍 VLM 机器人 agent 通过视觉反馈调用动作原语，但反复调用模型与冗余观测带来大量 token 开销。PyRUA-Lean 是交互式代码执行框架，将反馈驱动的原语组合与精简观测结合，成功率提升 14% 的同时减少 65% token。
👍 27
High-resolution 3D generation increasingly relies on voxel latents and multi-stage pipelines that first predict active structure and then synthesize local geometry. While effective, this design fragments continuous surfaces into many local tokens, inflates generation cost, and often weakens topologi
中文介绍 高分辨率 3D 生成依赖体素隐变量与多阶段流程：先预测活跃结构再合成局部几何，易将连续表面碎片化、抬高生成成本并削弱拓扑一致性。SILSA 提出滑动窗口切片隐变量，实现保持拓扑的高分辨率 3D 生成。
👍 26
Furnished floor plans support real-estate visualization, interior design, and architectural workflows, yet automatic furnishing remains challenged by limited real-world data and the need to satisfy interacting geometric and functional constraints. We ask whether professional furnishing knowledge can
中文介绍 户型图自动家具布置可服务房产可视化、室内设计与建筑流程，但受限于真实数据稀缺以及几何与功能约束相互耦合。Architect-Ant 尝试利用专业布置知识，生成可编辑的家具布局，满足多重交互约束。
👍 25
Looped Transformers achieve parameter efficiency by repeatedly executing a shared block across recurrent loops. Each loop yields an intermediate representation decodable for the same next token, yet standard decoding discards earlier states. Because earlier loops embody less computation, recurrence
中文介绍 Looped Transformer 通过重复执行共享模块实现参数高效，每次循环都产生可解码下一 token 的中间表示，但标准解码丢弃早期状态。论文利用循环间计算量的差异改进解码策略，几乎不增加额外成本即提升生成效果。
👍 23
Group Relative Policy Optimization (GRPO) is widely used to train reasoning language models, where it computes advantages by centering and normalizing rewards across rollouts of the same prompt. For multiple rewards, GRPO sums the reward components and normalizes the total reward by its within-group
中文介绍 GRPO 在多奖励场景下直接求和并按组内方差归一化，忽略了各奖励之间的相关性。论文提出 CorrGRPO，以相关性归一化调整优势估计，使各奖励目标获得更均衡的学习信号，改善推理语言模型的多目标优化。
👍 23
We present Multimodal Flow, a fully continuous generative model of language and vision. Most unified multimodal models either model both language and quantized images as discrete tokens or combine discrete language prediction with continuous image generation. The former introduces a visual quantizat
中文介绍 提出 Multimodal Flow，在嵌入空间中对语言与视觉做全连续流建模。它规避了把图像量化为离散 token 带来的视觉量化误差，也无需「离散文本预测＋连续图像生成」的混合架构，实现统一的多模态生成。
👍 21
Video generation models are increasingly being explored as world simulators for embodied planning and learning. To do so effectively, these models must not only generate visually appealing frames, but also predict how environments dynamically evolve when executing goal-directed actions. While evalua
中文介绍 视频生成模型被视为具身规划与学习的世界模拟器，但现有评测多关注画面质量。Ego2Act 面向第一人称视频生成，评测模型能否预测执行目标导向动作后环境的动态演化，为具身世界模型提供更贴近落地的评价。
👍 21
Real-world enterprise data science and analytics workflows require reasoning across dozens of tables, performing statistical analyses, and acting on the results. Established text-to-SQL benchmarks evaluate query generation alone, and audits have found their answer keys frequently wrong. Because real
中文介绍 真实企业数据科学工作流需跨数十张表推理、执行统计分析并据此行动，而现有 text-to-SQL 基准仅评查询生成，其答案键常被审计发现错误。Argo-Bench 面向企业级工作流，评测数据 agent 的端到端能力。
👍 19
Extending a text embedding model to new modalities typically degrades text retrieval quality, and existing omni-modal embedders compensate with multi-billion parameters. We present Omni-Embed-Mini, a 0.9B-parameter model that maps text, speech, audio, images, video, and visually-rich documents into
中文介绍 把文本嵌入模型扩展到新模态常会损害文本检索质量，现有全模态嵌入模型动辄数十亿参数。Omni-Embed-Mini 仅 0.9B 参数，通过稠密蒸馏将文本、语音、音频、图像、视频与富视觉文档映射到统一空间且不遗忘文本能力。
👍 17
Vision-Language Models (VLMs) have shown strong multimodal reasoning capabilities, yet whether they truly capture the physical consistency underlying real-world dynamics remains unclear. Existing benchmark paradigms often suffer from fragmented evaluation, focusing on isolated cognitive stages while
中文介绍 现有 VLM 物理智能评测多聚焦孤立认知阶段，评测割裂，难以判断模型是否真正掌握真实世界动态的物理一致性。PhysVista 提出「感知—推理—评估」闭环基准，系统考察 VLM 的物理理解与推理能力。
👍 15
Preference distillation typically treats a teacher response as preferred and the student's own response as rejected. This assumes that self-generated failures are the most informative negatives and that rejects must come from a model at least as large as the student, making generation costly at scal
中文介绍 偏好蒸馏通常把教师回答当成正例、学生自身回答当成负例，默认失败样本最具信息量且负例模型不小于学生。该工作重新审视这一假设，研究负例来自更小模型的偏好蒸馏缩放规律，从而降低大规模生成成本。
👍 14
On-policy self-distillation has recently emerged as an effective approach for improving language-model reasoning by supervising students with a frozen or EMA version of themselves that receives privileged information. Its application to multimodal large language models (MLLMs), however, remains larg
中文介绍 on-policy 自蒸馏用冻结或 EMA 的自身副本携带特权信息来监督学生，可提升语言模型推理，但用于多模态大模型仍有限。Where-OPD 借助合成场景提供空间引导，将该方法扩展到 MLLM 的视觉定位与推理。
@jakeserval · 1.1K 粉丝 · 406.0K 阅 · 505 赞 · 1.1K 转
There is a lot of talk about the jobs AI takes away and very little about the ones it creates. While pessimism might be easier (and certainly gets more clicks), I’m more optimistic. From where I sit,
中文介绍 博主谈 AI 创造的新岗位：在唱衰 AI 抢工作的声浪之外，更关注被创造出来的职位，并重点讨论「自动化工程师」这一新兴角色的崛起逻辑。属于观点类长文，而非具体操作教程。
@0xROAS · 54.6K 粉丝 · 118.6K 阅 · 502 赞 · 56 转
Seedance 2.5 is one of the craziest "nano banana moments" of AI VIDEO in 2026. It's the model behind every viral video that you are seeing right now... and it's pretty simple to generate those video,
中文介绍 Seedance 2.5 被称为 2026 年 AI 视频的「nano banana 时刻」，是当下大量爆款视频背后的模型。博主给出用它生成这类病毒式短视频的具体流程，并强调上手门槛并不高。
@0xSigil · 56.0K 粉丝 · 79.1K 阅 · 509 赞 · 72 转
6 months ago, the #1 AI was Claude Opus 4.6. Our model, Underdog 27B beats it while running on your Mac or iPhone. Most AI will run on your device, not in data centers. Intelligence that recently
中文介绍 博主称自研的 Underdog 27B 模型跑在本地 Mac 或 iPhone 上，即可击败半年前的 Claude Opus 4.6，主张未来多数 AI 会运行在个人设备而非数据中心，强调「拥有自己的 AI」而不是被它控制。
@UberEng · 66.9K 粉丝 · 54.0K 阅 · 529 赞 · 59 转
Introduction Uber’s rapid adoption of AI agents has fundamentally changed how teams interact with code, data, and operational systems. Early ad-hoc integrations with the MCP (Model Context Protocol)
中文介绍 Uber 工程团队分享 MCP Gateway：AI agent 快速普及后，早期零散的 MCP（Model Context Protocol）集成难以为继，于是搭建统一的 MCP 管理平台。属于大厂基础设施实践复盘，适合关注 agent 工程化的人。
@VibeMarketer_ · 37.8K 粉丝 · 31.4K 阅 · 504 赞 · 59 转
i'm going to show you how to build a content machine with Opus 5.5 that turns your ideas, research and expertise into articles, posts and newsletters in your own voice. from finding your next topic to
中文介绍 一份完整课程：用 Opus 5.5 搭建内容生产机器，把想法、调研和个人专长转化成保持自己语气风格的文章、社媒帖与 newsletter，覆盖从选题挖掘到成稿发布的全流程。
@goyal__pramod · 11.4K 粉丝 · 24.7K 阅 · 504 赞 · 59 转
Now I am aware you must have been looking forward to the second part of my CUDA blog (if you haven't read it yet, check it out!), but hey! A man can have varied interests. And I believe if you are
中文介绍 博主开新坑，从零讲强化学习系列第一篇。作者此前写 CUDA 博客，这次转向 RL，面向想系统入门强化学习基础的读者，属于偏教学的长文连载，而非工具测评。
@ActionModelAI · 60.2K 粉丝 · 6.8K 阅 · 501 赞 · 439 转
OpenAI has chosen not to release GPT-6.1 Astra after the model failed to meet its safety standards. To be clear, GPT-6.1 Astra is not GPT-6.1 Sol, the model OpenAI released at DevDay. GPT-6.1 Sol has
中文介绍 OpenAI 因未达安全标准，决定不发布 GPT-6.1 Astra——它与 DevDay 上线的 GPT-6.1 Sol 并非同一模型。博主由此延伸讨论 AI 的权限与授权问题，属于新闻点评类内容。
中文介绍 Hugging Face 博客发布微软撰写的文章，题为「智能体说它已完成，数据库却不同意」，讨论智能体声称任务已完成、而实际系统状态并不一致的问题。
a quiet day.
中文介绍 Latent Space 的 AINews 栏目发布当日简讯，表示当天没有太多重要事件发生，整体是较为平静的一天。
Learn how startups can choose GPT-6 models, tune reasoning effort, improve prompts and skills, coordinate tools, and prepare workflows for production.
中文介绍 OpenAI 发布 GPT-6 系列模型使用指南，介绍创业公司如何选择 GPT-6 模型、调节推理强度、优化提示与技能、协调工具调用，并为生产环境准备工作流。
Enterprise AI is no longer a future ambition. It is in full operational flight. Model capabilities are advancing faster than most organizations can absorb, while the cost of performance continues to fall. Globally, AI investment is set to reach $2.5 trillion in 2026, up 44% from the previous year. F
中文介绍 MIT 科技评论文章指出，企业 AI 已进入实际运营阶段，模型能力提升快于多数组织的吸收速度，而性能成本持续下降；2026 年全球 AI 投资预计达 2.5 万亿美元，同比增长 44%。
中文介绍 Allen AI 在 Hugging Face 博客宣布开源 AstaBrief，这是其 Asta 系统中用于快速生成报告的模型。
After leading Meta’s Llama models, Ahmad Al-Dahle is now transforming Airbnb with AI — from how its teams develop products to how it serves guests.
中文介绍 曾主导 Meta Llama 系列模型的 Ahmad Al-Dahle 现已加入 Airbnb 推动 AI 转型，涉及团队内部的产品开发方式以及面向房客的服务体验。
On an afternoon in Seoul in March 2016, I watched a program I helped build put a stone on the fifth line of a Go board in what looked like a gift to its human opponent. Move 37 in game two of the five-game match looked so absurd that some commentators thought it was a…
中文介绍 MIT 科技评论文章以 2016 年 3 月首尔人机围棋赛第二局、被视作「送礼」的第 37 手为例，指出大语言模型并不真正具备推理能力。
the minimalist harness goes stable... and TypeScript!
中文介绍 Latent Space 的 AINews 报道 Pi 1.0 与 Pi Durable 发布，极简智能体框架进入稳定阶段并支持 TypeScript，同时提及 AIE NYC 活动。
中文介绍 ServiceNow AI 团队在 Hugging Face 博客介绍 AutoSynthData，一种为企业级智能体生成训练数据的方法。
We catch up with RLM first author Alex Zhang, MIT PhD, on Jev, PhD masxing, and the future of harnesses.
中文介绍 Latent Space 对话 RLM 论文第一作者、MIT 博士生 Alex Zhang，谈及 Jev、读博经历以及智能体框架的未来发展。
Chatham Financial uses Codex and GPT-5.6 to build technology and redesign workflows, cutting trade validation from 30 minutes to under 4.
中文介绍 Chatham Financial 使用 Codex 与 GPT-5.6 构建技术并重塑工作流程，将交易验证时间从 30 分钟缩短至 4 分钟以内。
中文介绍 TLDR AI 当日摘要涵盖三个话题：决策模型、被称为「Claude 式」的科学研究，以及 OpenAI 安全团队的人事解雇事件。
Advanced AI may matter most for the routine work behind breakthrough ideas. Explore why execution could shape the next economy and the pace of progress.
中文介绍 OpenAI 发文探讨先进 AI 对突破性想法背后常规工作的价值，认为执行环节可能决定下一阶段经济的形态与进步速度。
Albertsons Cos. is using ChatGPT Enterprise and the OpenAI API to help teams work faster and make grocery shopping easier for millions of customers.
中文介绍 Albertsons 公司正利用 ChatGPT Enterprise 与 OpenAI API 帮助团队提升工作效率，并为数百万顾客简化购物流程。
... but you can’t try it yet unless you are “government users and trusted cyber defenders in the Fairwind Program”
中文介绍 Latent Space 的 AINews 报道 Gemini 4 Argon，被视作 Google DeepMind 对 Astra/Fable 的回应，支持 100 万 token 输出；目前仅限 Fairwind 计划中的政府用户与受信任网络防御者试用。
美股整体偏多但动量分化：QQQ 收 749.58、单日 +1.02%，RSI 65.4，MACD 8.2047 高于信号线 6.4876，距 52 周高仅 -0.66% 并呈多头排列；SPY 769.64 虽贴近 52 周高（-1.25%），但 MACD 0.8356 已跌破信号线 0.8931 出现死叉，RSI 54.5 居中。个股中 NVDA（5 日 +3.95%、RSI 62.9）与 MSFT（MACD 金叉、RSI 62.6）维持多头结构，而 AAPL、GOOGL、TSLA、META 均录得 MACD 死叉，GOOGL RSI 恰为 50、价格夹在 SMA20 与 SMA50 之间，方向不明。加密板块仍普遍多头排列：BTC 84796.03、RSI 63.8，ETH 2690.38、RSI 60.8，SOL 119.96、RSI 64.1，但三者 MACD 均现死叉迹象，距 52 周高分别为 -32.81%、-43.42%、-49.45%；配合恐慌贪婪指数 65（贪婪）与总市值 2.90T USD（24h -2.14%）、BTC 主导率 58.6%、ETH 11.3%，情绪偏热而市值回落，呈价量背离特征。中概股为当日最弱技术面：BABA RSI 35.6、PDD 33.4、JD 30.4、0700.HK 40.6，全线 MACD 死叉并呈空头排列，0700.HK 更接近 52 周低。商品外汇方面，黄金 GC=F RSI 34.1、5 日 -3.68% 且空头排列；原油 CL=F 虽多头排列，但 MACD 0.5518 低于信号线 1.8207；美元指数 101.92 RSI 71.7 超买、距 52 周高 -0.28%，USDCNY 6.70 RSI 34.5 贴近 52 周低。宏观端 10Y 美债收益率 5.28% RSI 74.6 超买，VIX 15.31 单日 -6.59% 且 MACD 金叉。
收 749.58，单日 +1.02%、5 日 +0.68%，距 52 周高仅 -0.66%。RSI 65.4 位于中性偏强区，MACD 8.2047 显著高于信号线 6.4876，柱状体为正；价格同时高于 SMA20 727.78、SMA50 716.82 与 SMA200 668.6，呈标准多头排列，趋势字段 bullish，动量与均线结构同向。
收 233.95，单日 +1.34%、5 日 +3.95%，距 52 周高 -1.65%。RSI 62.9，MACD 3.4886 高于信号线 2.6614，柱状体为正；价格高于 SMA20 223.79、SMA50 218.28 与 SMA200 200.71，多头排列完整，5 日涨幅在样本中居前，短期动量读数偏强。
收 343.5，单日 +1.56% 但 5 日 -0.12%，距 52 周高 -15.93%。RSI 恰好 50，为全样本最中性的读数；MACD -0.5463 低于信号线 -0.3972 并已死叉，柱状体为负。价格 343.5 夹在 SMA20 342.62 与 SMA50 344.37 之间，趋势字段 neutral，方向信号互相抵消。
收 105.85，单日 -1.49%、5 日 -3.54%，距 52 周高 -44.17%。RSI 35.6 偏弱，MACD -2.3886 低于信号线 -2.1837 并已死叉，柱状体为负。价格低于 SMA20 110.07、SMA50 116.87 与 SMA200 130.93，趋势字段 bearish，均线空头排列，各项读数一致偏弱。
收 4162.3，单日 -0.95%、5 日 -3.68%，距 52 周高 -25.49%。RSI 34.1 属本组偏低读数，MACD -63.4969 明显低于信号线 -38.6231，柱状体为负。价格低于 SMA20 4334.33、SMA50 4366.27 与 SMA200 4553.37，呈空头排列，趋势字段 bearish，短期下行斜率明显。
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
以上内容仅为技术指标读数的客观描述，不构成投资建议或走势预测；过去走势不代表未来表现，指标存在滞后与失真可能，仅供技术指标解读参考。
Michael Kotlikoff described the allegations of a woman who says she was drugged and gang raped as "deeply disturbing".
中文摘要 康奈尔大学校长迈克尔·科特利科夫就一起兄弟会宿舍强奸指控表示，校方「必须做得更好」，并称一名女性自称被下药后遭轮奸的指控「令人深感不安」。
New York City Mayor Zohran Mamdani shared a video riding the subway with Irish PM Micheal Martin during UN week.
中文摘要 纽约市长佐赫兰·马姆达尼与爱尔兰总理米歇尔·马丁在联合国大会周期间一同乘坐地铁并玩宾果游戏，马姆达尼发布了相关视频。
Latest polls suggest Lula's lead over Bolsonaro is narrowing, with neither candidate likely to win outright on Sunday.
中文摘要 巴西大选前，卢拉与博索纳罗结束竞选造势。最新民调显示卢拉对博索纳罗的领先优势收窄，两人在周日投票中均不太可能直接胜出。
Fighting in Yemen displaces more civilians, as Trump says a decision on Iran is coming.
中文摘要 也门武装力量袭击萨那，当地战事导致更多平民流离失所；美国总统特朗普警告德黑兰称，有关伊朗的决定即将作出，伊朗将面临「艰难方式」。
Ghana is a predominantly Christian country where preaching on public transport is widespread.
中文摘要 加纳是以基督教为主的国家，在公共交通工具上传教十分普遍，但巴士上的牧师布道可能很快将难以见到或听到。
Prime Minister Mark Carney has said the military is running scenarios in the unlikely case the US invades Canada.
中文摘要 加拿大总理马克·卡尼表示，鉴于美国入侵加拿大虽属不太可能发生的情况，军方仍在就此进行情景推演，加方正为此做准备。
The US coastguard says it has dispatched air and surface crews to search for the plane near Nantucket.
中文摘要 一架空中救护飞机在从百慕大飞往波士顿途中失踪，美国海岸警卫队表示已派出空中和水面搜救力量，在楠塔基特附近海域搜寻该机。
Connie Chan, endorsed by Nancy Pelosi to succeed her in California, says she will vote to 'end a genocide in Gaza'.
中文摘要 获佩洛西背书、拟在加州接替其国会席位的候选人康妮·陈呼吁对以色列实施「全面武器禁运」，并称将投票以「结束加沙的种族灭绝」。
Man charged with nine offences after vehicle struck Knights fans attending grand final farewell, leaving 10 injured including three children Get our breaking news email, free app or daily news podcast Ten people have been injured, one critically, after a car crashed into a crowd of people in Newcast
中文摘要 澳大利亚纽卡斯尔一辆汽车撞入参加NRL总决赛告别活动的骑士队球迷人群，造成10人受伤，其中一名5岁女童昏迷；一名持P牌驾驶者被控九项罪名。
President Trump says he's been having success winning races in Latin America. As Brazilian voters go to the polls, they're also deciding how much influence Trump should have in their country.
中文摘要 美国国家公共电台报道，部分巴西选民将此次大选视为对特朗普的一次投票。特朗普称其在拉美选举中屡有收获，而巴西选民也在决定美国应有多大影响力。
High school students say their classrooms are overcrowded, teachers are missing and buildings need repairs. The protests have spread to cities across France, and some 5,000 people have been arrested.
中文摘要 法国高中生抗议教室拥挤、师资不足和校舍失修，示威蔓延至全国多座城市，约5000人被捕，其中多数为青少年。
Twenty-three years after President George W. Bush's campaign of "shock and awe," the U.S. military has completed its withdrawal from Iraq.
中文摘要 在时任美国总统乔治·W·布什发动「震慑与畏惧」军事行动23年后，美国军方已完成从伊拉克的撤军。
Brazil's election enters its final day of campaigning as Lula and Flávio Bolsonaro make their last pitches to voters. Brazilians explain what worries them, and why some still can't choose.
中文摘要 巴西大选进入最后一天竞选，卢拉与弗拉维奥·博索纳罗作最后拉票；部分选民表示两人都不想选，并说明了自己的担忧。
The aircraft lost communication with flight controllers after significantly dropping in altitude, according to flight data.
中文摘要 一架载有6人的医疗飞机在美国马萨诸塞州海岸附近失踪；飞行数据显示，该机在高度大幅下降后与空中交通管制失去联系。
Israeli settlers attack Palestinian farmers during olive harvest
中文摘要 半岛电视台发布视频显示，以色列定居者在橄榄收获季节袭击巴勒斯坦农民。
The US and Mexico have jointly deployed military personnel in a border area in Texas as part of “mirror operations,” the Mexican government said Saturday.
中文摘要 墨西哥政府10月3日表示，美国与墨西哥已在得克萨斯州边境地区联合部署军事人员，作为双方「镜像行动」的一部分。
President Luiz Inácio Lula da Silva and right-wing challenger Flávio Bolsonaro remain locked in a tight race as Brazil prepares to vote in Sunday’s election, according to a pair of new polls.
中文摘要 巴西大选投票前夕，两项最新民调显示，总统卢拉与右翼挑战者弗拉维奥·博索纳罗支持率仍然持平，选情极为胶着。
Borrowing is getting harder for companies.
中文摘要 彭博信用周报指出，派拉蒙债务下跌给借款人带来麻烦，企业借贷正变得愈发困难。
The news doesn’t stop when markets close. Hosts David Gura, Christina Ruffini and Lisa Mateo bring clarity, context and a bit of humor to the weekend’s biggest headlines, LIVE from New York. Joined by The Atlantic Staff Writer Vivian Salama, “The Profiler” Online Job Scam Hunter Jay Jones, UMich Pro
中文摘要 彭博周末新闻节目由戴维·古拉、克里斯蒂娜·鲁菲尼和莉萨·马特奥在纽约直播主持，梳理周末重大头条，并邀请《大西洋月刊》撰稿人薇薇安·萨拉马及网络求职骗局追踪者等嘉宾参与。
Flight 1073 was heading to Tel Aviv when the attack was launched in the cockpit
中文摘要 据《金融时报》报道，迪拜航空1073航班在飞往特拉维夫途中，副驾驶在驾驶舱内以斧头实施「恐怖主义行为」。
Bloomberg Pursuits reporter Sarah Rappaport tells Bloomberg This Weekend that luxury travelers are increasingly seeking tiny hotels with as few as one to six rooms, combining the privacy of a villa with personalized high-end hotel service. Speaking with hosts David Gura and Christina Ruffini, Rappap
中文摘要 彭博报道称，越来越多高端旅客青睐客房仅一至六间的微型酒店，这类酒店兼具别墅的私密性与个性化的高端酒店服务。
Pointed offers a strategic twist to the news quiz format, testing not just players’ knowledge of the news but also their confidence in their answers. Join Bloomberg's Lisa Mateo, Christina Ruffini and David Gura as they play and check out the quiz for yourself at Bloomberg.com (Source: Bloomberg)
中文摘要 彭博推出策略型新闻测验节目「Pointed」，不仅测试参与者对新闻的掌握，也考验其对答案的信心，由莉萨·马特奥、克里斯蒂娜·鲁菲尼和戴维·古拉主持参与。
People starting to speak like AI chatbots, AI anxiety spilling into therapists' offices, NYC's AI office boom a bust for young job hunters and Tinder rolls out group dating for younger users. Join Lisa Mateo, David Gura and Christina Ruffini for a roundup of headlines you may have missed, but gotta
中文摘要 该节目盘点易被忽略的头条：人们开始像人工智能聊天机器人一样说话、人工智能焦虑进入心理诊室、纽约市人工智能办公热潮令年轻求职者失望，以及Tinder面向年轻用户推出群体约会。
Washington Institute research director Dana Stroul tells Bloomberg This Weekend that the US military buildup in the Middle East, including a third aircraft carrier, is expanding President Donald Trump’s options as diplomacy with Iran remains stalled and Washington weighs its approach to the Houthis
中文摘要 华盛顿研究所研究主任达娜·斯特劳尔表示，美国在中东的军事集结（包括部署第三艘航母）扩大了特朗普的选择空间，而美方与伊朗的外交仍陷于停滞。
Fraud investigator Jay Jones, known online as “The Profiler,” tells Bloomberg This Weekend that AI is helping scammers create increasingly convincing fake recruiters and job listings that can lead victims to surrender money and sensitive personal information. Speaking with hosts David Gura and Chris
中文摘要 欺诈调查员杰伊·琼斯（网名「侧写师」）表示，人工智能正帮助诈骗者制作愈发逼真的虚假招聘人员和职位信息，导致求职者损失钱财并泄露敏感个人信息。
Atlantic Staff Writer Vivian Salama tells Bloomberg This Weekend that Egypt’s then-intelligence chief Abbas Kamel made an unusual 67-minute trip to Israel less than two weeks before the Oct. 7, 2023, attack to warn Israeli officials that intelligence indicated Hamas was mobilizing for a major assaul
中文摘要 《大西洋月刊》撰稿人薇薇安·萨拉马表示，埃及时任情报负责人阿巴斯·卡迈勒曾在2023年10月7日袭击前不到两周，罕见地对以色列进行67分钟访问，警告以方情报显示哈马斯正在动员。
Debate over artificial intelligence has largely focused on whether machines could become too powerful. Harvard philosopher Michael Sandel argues that a more immediate question is whether AI changes humans themselves: weakening authentic connection, independent thought and the ways people understand
中文摘要 哈佛哲学家迈克尔·桑德尔认为，关于人工智能的争论多集中于机器是否变得过于强大，但更紧迫的问题是人工智能是否正在改变人类自身，削弱真实连接与独立思考。
6 回复 · 程序员 节点
8 回复 · 程序员 节点
5 回复 · 程序员 节点
6 回复 · Apple 节点
16 回复 · 程序员 节点
11 回复 · 程序员 节点
9 回复 · 程序员 节点
29 回复 · Apple 节点
13 回复 · Apple 节点
17 回复 · Apple 节点
该源今日无内容。