Python · ★ 28,784 · 🍴 2,438 · 📈 439 stars today
AI agent skill that researches any topic across Reddit, X, YouTube, HN, Polymarket, and the web - then synthesizes a grounded summary
中文介绍 面向 AI Agent 的信息研究技能插件，能自动跨 Reddit、X（Twitter）、YouTube、Hacker News、Polymarket 及通用网页抓取任意主题的近期内容，并将多源信息综合归纳成有据可查的摘要。适合需要快速掌握某话题最新动态的研究人员和自动化工作流开发者，无需手动搜索多个平台。
TypeScript · ★ 33,199 · 🍴 4,239 · 📈 631 stars today
The Frontend Stack for Agents & Generative UI. React, Angular, Mobile, Slack, and more. Makers of the AG-UI Protocol
中文介绍 面向 AI Agent 与生成式 UI 的前端开发框架，支持 React、Angular、移动端及 Slack 等多平台接入。提出并维护 AG-UI 协议，帮助开发者将 AI 能力无缝嵌入应用界面，简化 Copilot 功能的集成流程。适合希望在产品中快速构建 AI 助手交互体验的前端工程师。
Python · ★ 54,272 · 🍴 7,109 · 📈 446 stars today
The best-benchmarked open-source AI memory system. And it's free.
中文介绍 基准测试评分领先的开源 AI 记忆系统，免费使用。专为大型语言模型提供持久化、可检索的记忆能力，解决 LLM 上下文窗口有限、无法跨会话记忆的问题。适合构建长期对话助手、个人知识管理工具或需要记忆增强的 AI 应用的开发者。
TypeScript · ★ 14,955 · 🍴 2,123 · 📈 70 stars today
Agentic AI Infrastructure for magnifying HUMAN capabilities.
中文介绍 以 Agentic AI 为核心的个人 AI 基础设施搭建方案，目标是通过自动化 Agent 流水线放大个人的认知和执行能力。提供可自部署的工具链配置，适合技术型个人用户、研究员及希望构建专属 AI 工作流的知识工作者，用于信息处理、决策支持等场景。
JavaScript · ★ 1,765 · 🍴 256 · 📈 213 stars today
OpenAI Plugins
中文介绍 OpenAI 官方插件仓库，集中托管和发布 ChatGPT 插件相关规范、示例及文档。开发者可参考此仓库构建与 ChatGPT 集成的第三方插件，扩展模型的外部调用能力，适合希望将自有服务接入 ChatGPT 生态的开发团队和独立开发者。
Python · ★ 22,309 · 🍴 1,907 · 📈 683 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 为 AI Agent 提供「互联网之眼」的命令行工具，支持读取和搜索 Twitter、Reddit、YouTube、GitHub、Bilibili、小红书等主流平台内容，零 API 费用即可使用。通过统一 CLI 接口屏蔽各平台差异，适合需要让 Agent 获取多平台实时信息的开发者和自动化脚本场景。
JavaScript · ★ 86,981 · 🍴 4,937 · 📈 25 stars today
web development for the rest of us
中文介绍 面向大众的现代 Web 前端框架，采用编译时方案将组件代码转换为高效原生 JavaScript，无需运行时虚拟 DOM 开销，包体积更小、性能更优。适合追求简洁开发体验和高性能页面的前端开发者，尤其适用于中小型 Web 应用和交互组件开发。
C · ★ 30,688 · 🍴 7,956 · 📈 20 stars today
The official NGINX Open Source repository.
中文介绍 NGINX 官方开源仓库，提供高性能 HTTP 服务器、反向代理、负载均衡及邮件代理功能。以事件驱动架构著称，能以极低内存处理高并发连接。广泛用于生产环境的 Web 服务器部署、静态资源分发、API 网关及微服务流量路由等场景，是运维工程师的核心基础设施组件。
Go · ★ 35,999 · 🍴 454 · 📈 159 stars today
Find vulnerabilities, misconfigurations, secrets, SBOM in containers, Kubernetes, code repositories, clouds and more
中文介绍 全面的云原生安全扫描工具，可检测容器镜像、Kubernetes 集群、代码仓库、云基础设施中的漏洞、配置错误、敏感信息和软件物料清单（SBOM）。支持多种扫描目标和 CI/CD 集成，适合 DevSecOps 团队在开发和部署各阶段进行安全合规审查。
Go · ★ 134,505 · 🍴 19,085 · 📈 30 stars today
The Go programming language
中文介绍 Go 编程语言的官方源码仓库，由 Google 主导维护。Go 以静态类型、编译速度快、原生并发支持（goroutine/channel）和简洁语法著称，适合构建高性能服务端应用、云原生工具链、微服务及命令行工具，是后端工程师和 DevOps 领域广泛采用的主流语言。
TypeScript · ★ 26,604 · 🍴 3,037 · 📈 794 stars today
An Open Source implementation of Notebook LM with more flexibility and features
中文介绍 Google NotebookLM 的开源替代实现，在功能和灵活性上进行了扩展。支持本地部署，用户可导入文档、网页等多种资料并借助 LLM 进行问答、摘要与知识整理。适合重视数据隐私、希望自建 AI 知识笔记本的研究人员、学生及知识工作者。
Shell · ★ 219,647 · 🍴 19,542 · 📈 700 stars today
An agentic skills framework & software development methodology that works.
中文介绍 面向软件开发的 Agentic 技能框架，提供一套实际可用的 Agent 技能组织方法论。帮助开发者将复杂开发任务分解为可由 AI Agent 执行的技能模块，提升 AI 辅助编程的可靠性和可控性。适合探索 AI 驱动软件工程工作流的开发者和技术团队。
JavaScript · ★ 49,332 · 🍴 10,197 · 📈 193 stars today
AI-powered job search system built on Claude Code. 14 skill modes, Go dashboard, PDF generation, batch processing.
中文介绍 基于 Claude Code 构建的 AI 求职自动化系统，集成 14 种技能模式，涵盖职位匹配、简历生成（PDF 输出）、批量申请处理等功能，并配套 Go 语言开发的数据看板。适合正在密集求职的技术人员，通过 AI 流水线显著提升求职效率和投递质量。
Python · ★ 101,854 · 🍴 12,440 · 📈 150 stars today
Robust Speech Recognition via Large-Scale Weak Supervision
中文介绍 OpenAI 开源的通用语音识别模型，基于大规模弱监督训练，具备强健的多语言转录和翻译能力。支持多种模型尺寸，在噪声环境和口音场景下表现优异。适合字幕生成、会议记录、语音助手等需要高精度语音转文字的开发者和研究人员。
TypeScript · ★ 81,178 · 🍴 8,274 · 📈 25 stars today
Next generation frontend tooling. It's fast!
中文介绍 下一代前端构建工具，利用原生 ES 模块实现极速冷启动和热模块替换（HMR），生产环境使用 Rollup 打包输出优化产物。相比 Webpack 等传统工具显著缩短开发反馈周期，支持 Vue、React、Svelte 等主流框架，已成为现代前端项目的主流脚手架选择。
Rust · ★ 573 · 🍴 24 · 📈 64 stars today
Policy-driven, layered isolation and containment
中文介绍 微软开源的策略驱动、多层次隔离与容器化框架，通过声明式策略实现细粒度的资源隔离和访问控制。适合需要在多租户或安全敏感环境中对工作负载进行严格隔离的云平台工程师和系统架构师，可用于强化容器和进程级别的安全边界。
Python · ★ 80,956 · 🍴 10,657 · 📈 433 stars today
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
中文介绍 基于 PaddlePaddle 框架的轻量级多语言 OCR 工具包，支持 100 余种语言，可将 PDF 和图片文档转换为结构化数据，充当图像/PDF 与 LLM 之间的桥梁。适合文档智能处理、数据提取、RAG 系统预处理等场景，广泛用于企业文档数字化和 AI 应用数据接入。
Python · ★ 48,470 · 🍴 5,391 · 📈 216 stars today
Open-Source Frontier Voice AI
中文介绍 微软开源的前沿语音 AI 项目，专注于构建高质量的开源语音交互能力。适合需要语音合成、语音对话或语音界面的开发者，可用于构建智能助手、无障碍工具及语音驱动的应用，降低企业和个人接入先进语音 AI 技术的门槛。
👍 1
In this paper, we study regret minimization in repeated games with adaptive opponents who can respond based on histories of play. The standard metric of external regret in online learning is known to fail to capture such adaptivity. To account for players' counterfactual reasoning, we introduce {\tt
👍 3
Multimodal Large Language Models (MLLMs) excel at 2D semantic understanding but lack intrinsic 3D awareness, resulting in representations that fail to maintain geometric and spatial consistency across video frames. Given the scarcity of large-scale 3D data, we present GeoVR, a novel framework that l
👍 7
Vision-Language-Action (VLA) models leverage the rich world knowledge of pretrained vision-language models (VLMs) to enable instruction-following robotic manipulation. However, the structural mismatch between VLM semantic spaces and embodied control policies often hinders the learning of precise per
👍 63
Code language models need repository-level context to resolve imports, APIs, and project conventions. Existing methods inject this knowledge as long inputs (retrieved through RAG or dependency analysis) or through per-repository fine-tuning and LoRA -- costly at repository scale and brittle to evolv
👍 1
A situated query like "where is Lin Wei?" often encodes more than its literal content: the user may also want to know whether Lin Wei is free, in a good mood, or worth interrupting now. Standard tool-use agents answer the literal question and stop. AURA inserts an inference step between scene percep
👍 2
Benchmarks are fundamental for evaluating and advancing LLMs and MLLMs by providing standardized and explicit measures of performance. However, their construction is labor-intensive and hard to reuse, raising concerns about sustainability and scalability. Moreover, existing benchmarks often quickly
👍 2
Large language models are increasingly used to simulate social media users and infer how individuals may respond to online discussions. However, it remains unclear whether these simulations reflect precise user-specific beliefs or whether they are highly sensitive to semantically independent changes
👍 0
AI research often requires decisions before future evidence exists: which bottleneck to attack, which direction to pursue, or where a project should be positioned. We introduce ForeSci, a temporally controlled benchmark for evaluating whether LLM agents can make such forward-looking research judgeme
👍 15
Video generation models have made impressive strides in synthesizing visually compelling content, yet their outputs remain confined to the virtual domain. A natural question follows: how well do these models reflect the physical world when their generated videos leave the screen and enter reality? W
👍 4
Temporal Grounding (TG) aims to localize video segments corresponding to a textual query. Prior research predominantly focuses on single-segment retrieval. Real-world scenarios, however, often require localizing multiple disjoint segments for a single query -- a setting we term One-to-Many Temporal
👍 7
Large language models can reproduce training data, but existing memorization evaluations mostly measure whether models can be forced to do so, rather than whether they do so under ordinary use. We introduce PropMe, a propensity-aware framework for memorization evaluation that contrasts prefix-based
👍 36
Planning for real-world problems by language models often involves both world and user constraints, which may not be fully specified upfront and are progressively disclosed through interaction. However, existing benchmarks still underexplore adaptive planning under such progressively revealed dual c
👍 42
Role-playing language agents (RPLAs) should play characters whose values and behavior evolve as the story progresses, not maintain a fixed persona. Existing benchmarks measure factual recall at a given chapter, not whether responses align with the character's psychological trajectory, especially in
👍 23
Prior work has shown that large language models (LLMs) can translate unseen or low-resource languages by undergoing continued training or even by encoding a grammar book in their context. However, both methods typically overfit specific languages, with limited zero-shot transfer at test time. To tra
👍 38
Agents are widely deployed as assistants over documents, tools, and code. However, they typically act only on explicit user requests, which surface only the problems the user has noticed, while many other important problems coexist, hidden in plain sight, within the broader user context, with their
👍 8
Selection is a core operation in interactive image editing. To be practical, a user should be able to specify and disambiguate the desired selection region through either text or click-based interactions, and the system should support selecting not only objects but also other criteria, such as mater
👍 4
In robotics systems, vast amounts of visual data are easily captured at high resolution using low-cost, low-power hardware. Yet, limited bandwidth and on-device compute resources prevent full utilization when transmitted via conventional codecs like JPEG/MPEG. Newer codecs, like AV1/AVIF, improve th
👍 23
While household robots are often evaluated based on task completion, everyday domestic environments involve value-conflicting situations in which robots are expected to choose actions that prioritize other values than task success, such as human autonomy, efficiency, or social appropriateness. Yet,
👍 2
Recent progress in Large Language Model (LLM) agents has enabled promising advances in automated data science. However, existing approaches remain fundamentally limited by their static action sets and lack of principled long-horizon context management, hindering their ability to accumulate reusable
👍 1
Agentic LLMs with web search change the threat model for text anonymization: weak contextual cues can become cross-referenceable evidence for re-identification, yet those same details also carry downstream analytic value of the text. Existing defenses either remove explicit identifiers, perturb text
👍 2
Financial AI agents often fail for a simple reason: they make users carry the complexity. A user must repeatedly restate goals, risk preferences, portfolio context, past judgments, and shifting market assumptions, while the agent answers, retrieves, acts, and forgets. In finance, this is not just in
👍 4
Video is temporally redundant: adjacent frames usually share most objects, background, and layout. Yet existing video multimodal large language models (video MLLMs) usually encode each sampled frame as an independent RGB image, causing visual tokens to repeat content already present in earlier frame
👍 0
Large language models are increasingly deployed as coding agents, shifting safety from individual responses to action sequences. Existing benchmarks, however, primarily assess whether models refuse unsafe prompts, leaving impacts on stateful workspaces largely unexamined. We present SABER, a benchma
👍 3
Large Language Models exhibit paradoxical fragility in fundamental arithmetic, implying a disconnect between internal computation and discrete output. By analyzing the residual stream geometry during multi-operand addition, we identify the Iso-Raw-Sum Trajectory (IRST), a geometric structure where r
👍 3
Multimodal Large Language Models (MLLMs) have demonstrated significant achievements in general visual question answering (VQA) tasks. However, they remain brittle on mechanical engineering drawings, where high annotation density and weak domain knowledge, compounded by unreliable spatial relation re
👍 1
Reinforcement Learning with Verifiable Rewards (RLVR) has recently emerged as the cornerstone for shaping the remarkable coding abilities of Large Language Models (LLMs). However, the scalability of RLVR is severely constrained by the scarcity of sufficiently challenging verifiable code tasks that t
👍 1
Music recommendation systems typically treat songs as opaque tokens, relying on collaborative interaction histories which overlooks semantic or acoustic content. Prior work has explored LLM-augmented, multimodal, and text-enhanced approaches to sequential recommendation, and while some methods parti
👍 6
Memory-augmented LLM agents tackle complex long-horizon tasks by recursively summarizing interaction trajectories into compact memory. However, existing approaches typically train these memory policies using outcome-based reinforcement learning, failing to localize where intermediate memory quality
👍 3
Off-policy reinforcement learning of pretrained flow policies remains challenging due to the instability of optimization arising from the multi-step sampling process. Recently, Q-learning with Adjoint Matching (QAM) addressed this issue by reformulating into a memoryless stochastic optimal control (
👍 1
Diffusion-based image editing has achieved strong visual fidelity under natural language instructions, yet most existing systems still operate at the level of surface instruction following, without reasoning about the implicit contextual constraints embedded in real user requests. This often leads t
@DamiDefi · 96.5K 粉丝 · 2.3M 阅 · 584 赞 · 80 转
The number that stopped me was not the $2 trillion valuation. It was $791 million. That is what SpaceX made in net income in 2024. A profitable, growing aerospace company with a genuine moat in launch
中文介绍 博主将 SpaceX 招股说明书（S-1，约300页）喂给 Claude 进行分析，重点挖掘出财务数据亮点：SpaceX 2024年净利润达7.91亿美元，是一家有真实护城河的盈利航天公司。帖子以「$791 million」这一数字为切入点，梳理文件中被忽视的关键信息，为关注 SpaceX IPO 的投资者提供 AI 辅助阅读示范。
@sairahul1 · 110.7K 粉丝 · 710.8K 阅 · 509 赞 · 97 转
How To Become An AI Engineer in 2026. Without a CS degree. Without a bootcamp. Without knowing what a transformer is today. Here's what nobody tells you: The companies hiring right now don't need
中文介绍 分享2026年无CS学位转型AI工程师的路径建议。核心观点是：当前招聘市场不要求候选人精通Transformer原理，而是看重能用AI工具解决实际问题的能力。帖子针对非科班背景人群，提供绕开传统教育路径（学位/训练营）进入AI行业的实操思路。
@maubaron · 16.9K 粉丝 · 233.8K 阅 · 506 赞 · 19 转
Our YouTube channel has 125k subscribers and we've never made or uploaded a single video ourselves. This is a completely automated system. It is this very same strategy that made us the first app
中文介绍 介绍一套全自动YouTube运营系统：博主从未亲自制作或上传视频，频道已积累12.5万订阅。该策略据称使其成为某类目「第一个应用此方法」的应用，帖子详细拆解这套自动化内容生产与发布流程，适合对AI批量生产视频内容感兴趣的创作者参考。
@intuitiveml · 6.4K 粉丝 · 171.3K 阅 · 524 赞 · 70 转
Most agent frameworks today assume a desktop. One user, one machine, one process. The agent runs while the laptop is open, writes to a local filesystem, holds API keys in environment variables, and
中文介绍 分享构建云端 Agent 基础设施的实践心得。核心挑战在于：现有 Agent 框架大多基于「单用户+本地机器+单进程」的桌面假设，而云端场景需要处理多用户并发、无本地文件系统、API 密钥管理等完全不同的约束。帖子总结了从桌面迁移到云端时架构层面的关键差异与踩坑经验。
@dkundel · 19.3K 粉丝 · 116.9K 阅 · 523 赞 · 40 转
We launched the goal mode (or /goal) as a way to help you have Codex drive towards a concrete outcome. When you set a goal Codex will continue to work until the goal is achieved, whether that takes
中文介绍 Codex 官方介绍新功能 /goal（目标模式）的使用指南。设定目标后，Codex 会持续自主工作直至达成该目标，无论耗时多长。帖子说明了 /goal 的设计意图、触发机制及适用场景，是 OpenAI Codex 产品的功能发布说明。
@itsreallyvivek · 3.6K 粉丝 · 65.8K 阅 · 521 赞 · 28 转
A few days ago I wrote that getting into a frontier AI lab mostly comes down to two things: proven research and trench engineering. The more I think about it, the less these feel like separate skills.
中文介绍 作者分享进入顶级 AI 实验室的个人观察与思考。此前认为关键在于「研究成果」与「工程实战」两项，但深入思考后发现二者并非独立技能，而是高度融合的。帖子偏向经验性反思，适合有意加入 Anthropic、OpenAI 等前沿 AI 机构的候选人参考。
@dickiebush · 441.8K 粉丝 · 57.7K 阅 · 519 赞 · 45 转
Legendary marketer David Ogilvy generated over $864 million for his clients. He was a British advertiser known as "The Father of Advertising." And in 1982, Ogilvy sent this 1-page memo to his staff:
中文介绍 博主将广告大师 David Ogilvy 1982年写给员工的写作备忘录输入 Claude，构建了一个模拟「广告之父」风格的 AI 写作教练。Ogilvy 生涯为客户创造超8.64亿美元价值，其写作原则至今被奉为经典。帖子展示如何用历史名人的一手文本为 Claude 注入专家人格，打造垂直写作助手。
@sheriyuo · 8.6K 粉丝 · 30.6K 阅 · 512 赞 · 44 转
After seeing several people receive PhD offers and then immediately land highly paid industry positions during spring recruiting, I started wondering whether going straight into industry might
中文介绍 作者观察到2026年春招中，多名拿到博士录取的候选人随即转向高薪业界岗位，由此引发对「直接入行 vs 读博」路径的思考，并整理了强化学习（RL）方向的面试题集。帖子兼具职业规划反思与技术备考资料双重属性，适合对 RL 方向求职感兴趣的读者。
中文介绍 该文章介绍了一个由五个实验室、五支团队协作的项目，探索如何利用小型语言模型构建多模型金融主题叙事应用，展示了小模型在垂直场景中的协同潜力。
a quiet day of RSI.
中文介绍 Latent Space每日AI资讯简报，当日AI领域无重大事件发生，整体处于平静状态，编辑以「RSI」一词形容当天的低活跃度。
Your broken harness is actively making the model worse. Here's what I keep seeing after years of eyeballing trajectories, and what you need to fix.
中文介绍 Latent Space发文探讨强化学习（RL）环境质量问题，作者指出低质量的训练环境会主动损害模型性能，并根据多年观察轨迹的经验，列举常见问题并提出改进建议。
On June 5, 404 Media reported that attackers had been using Meta’s AI customer support agent to steal Instagram accounts. Their approach was simple: They asked the agent to link the accounts to email addresses that they controlled, and the agent complied. One attacker broke into the dormant Obama Wh
中文介绍 据404 Media于6月5日报道，攻击者利用Meta的AI客服代理窃取Instagram账户，方法是直接要求该代理将账户关联至攻击者控制的邮箱，代理照单全收。此事件暴露了AI客服系统在身份验证与权限控制方面的安全漏洞，引发业界对AI安全机制的广泛讨论。
a quiet day
中文介绍 Latent Space每日AI资讯简报，当日AI领域无重大进展，为平静的一天。
**Anthropic's Mythos/Opus cycle** sparked mixed reactions with praise for **Claude Mythos**'s one-shot workflows and concerns over **Opus 4.8** benchmark regressions. **Opus 4.7** showed strong chemistry task performance, "making Claude a chemist." **Sakana AI** launched an **RSI Lab** focusing on r
中文介绍 Anthropic的Mythos/Opus系列引发分歧，Claude Mythos的一次性工作流受到好评，但Opus 4.8出现基准测试性能回归；Opus 4.7在化学任务上表现强劲。Sakana AI发布RSI（递归自我改进）相关项目。整体当日无重大突破性新闻。
中文介绍 TLDR AI 2026年6月5日简报涵盖：Anthropic疑似泄露新模型Oceanus相关信息、ChatGPT推出「Dreaming」记忆功能，以及AI递归自我改进（RSI）领域的最新进展。
We talk with the VendingBench authors on evaling Claudes from Haiku to Mythos, and how they build leading, and lasting, frontier evals from scratch.
中文介绍 Latent Space专访Andon Labs创始人Lukas Petersson与Axel Backlund，两人是VendingBench评测框架的作者，分享了如何对Claude从Haiku到Mythos系列进行系统性评估，以及如何从零构建具有前瞻性和持久性的前沿AI评测体系。
中文介绍 NVIDIA在Hugging Face发布Nemotron 3.5 Content Safety模型，定位为面向全球企业AI的可定制多模态内容安全解决方案，支持企业根据自身需求灵活配置安全策略。
中文介绍 ServiceNow AI在Hugging Face发布EVA-Bench Data 2.0评测数据集，覆盖3个领域、121种工具、213个测试场景，用于评估AI智能体在企业级任务中的工具调用与工作流执行能力。
Learn how Endava is using AI agents, ChatGPT Enterprise, and Codex to accelerate software delivery, automate workflows, and build an AI-native culture across the enterprise.
中文介绍 软件服务公司Endava与OpenAI合作，将AI代理、ChatGPT Enterprise及Codex整合进软件交付流程，以加速开发周期、自动化工作流，并在企业内部推进AI原生文化转型。
Most days in her chambers, Judge Maritza Braswell, a federal magistrate judge in Colorado, sifts through stacks of documents written by people without a lawyer. Many of them can’t afford to hire a lawyer, and others have cases too weak or too small to interest one. She reads each one carefully, mind
中文介绍 科罗拉多州联邦治安法官Maritza Braswell每天需审阅大量由AI生成的诉状，这些文件来自无力负担律师费或案件不足以吸引律师的当事人。AI生成诉讼文书的大规模涌入正给美国法院系统带来新挑战，引发对司法效率与文书真实性的担忧。
ChatGPT introduces a new memory system to better remember preferences, keeping context fresh and relevant across conversations.
中文介绍 OpenAI为ChatGPT推出名为「Dreaming」的新记忆系统，能够更好地记住用户偏好，并在跨对话中保持上下文的时效性与相关性，旨在提升ChatGPT的个性化服务能力。
**NVIDIA** released **Nemotron 3 Ultra**, a fully open **550B MoE** model with **55B active parameters** and **1M context**, optimized for long-running agent tasks with up to **5x speedup** and **30% cost reduction**. It features hybrid Mamba/attention, LatentMoE, native MTP, and was pretrained on *
中文介绍 NVIDIA发布Nemotron 3 Ultra全开放模型，参数规模为550B MoE架构，活跃参数55B，支持100万token上下文，专为长时间运行的智能体任务优化，推理速度提升最高5倍、成本降低30%。模型融合Mamba/注意力混合架构、LatentMoE及原生多token预测（MTP）等技术。
a quiet day.
中文介绍 Latent Space每日简报聚焦图像生成领域：Reve 2与Ideogram 4相继推出，两款产品均强化了图像布局控制能力，是当日AI图像生成领域的主要进展，整体为平静的一天。
本期技术面整体呈现风险资产普遍承压、避险与防御性资产相对强势的分化格局。美股方面，SPY 当日下跌 -2.58%，QQQ 单日重挫 -4.80%，均触发 MACD 死叉或动量走弱信号；TSLA 与 META 的 MACD 死叉叠加空头排列，技术结构明显偏弱；NVDA 单日跌幅达 -6.20%，价格跌破 SMA20（219.10）压力位，整体科技板块技术面趋于中性偏弱。MSFT 距 52 周高点已回落 -24.99%，趋势标注转为中性。加密货币板块技术面最为恶化：BTC RSI14 仅 15.4、ETH 仅 12.6，双双深陷超卖，SOL 同样 RSI14 为 15.5，三者均呈空头排列且逼近 52 周低点。加密恐慌贪婪指数报 12（极度恐慌），加密总市值 2.17 万亿美元、24 小时下滑 -0.79%，BTC 主导率 56.1% 显示资金持续向比特币相对集中但整体仍在撤出。中概股方面，阿里、拼多多、京东均处于空头排列，RSI14 均在 37-41 之间，技术弱势延续；腾讯控股触发 MACD 金叉但整体仍为空头排列，5 日涨幅 +6.09% 形成局部分歧。商品与外汇方面，黄金期货 RSI14 降至 34.2，价格跌破多条均线；WTI 原油 RSI43.4 偏中性。宏观层面，VIX 单日飙升 +39.68% 并触发 MACD 金叉，美债 10 年期收益率 4.54% 维持多头排列，美元指数接近 52 周高点并呈强势多头排列，宏观信号整体指向市场避险情绪显著抬升。
ETH 当前报价 1567.01 美元，5 日跌幅达 -21.78%，距 52 周高点已下行 -68.37%，RSI14 读数仅 12.6，处于深度超卖区间。价格运行于 SMA20（1980.67）、SMA50（2173.10）、SMA200（2456.64）三线全面下方，呈典型空头排列。MACD 为 -140.44，信号线为 -95.56，柱状图持续扩大负值，动量进一步走弱。技术结构全面偏空，RSI 极端超卖虽可能触发短期反弹，但趋势信号尚未转向。
VIX 当前报 21.51，1 日涨幅 +39.68%，5 日涨幅 +40.40%，显示市场恐慌情绪急剧升温。价格已突破 SMA20（17.12）、SMA50（18.99）及 SMA200（18.42），呈多头排列。MACD 金叉信号触发，MACD 为 -0.40 但信号线 -0.77，直方图由负转正动量向上。RSI14 达 65.7 接近高位区间。指标结构显示短期波动率仍具上行动能。
TSLA 当前报 391.00 美元，1 日跌幅 -6.56%，5 日跌幅 -10.28%，距 52 周高点回落 -21.62%。价格低于 SMA20（425.87）与 SMA200（414.14），呈空头排列。MACD 死叉已触发，MACD 读数 4.14，信号线 8.61，负向动量持续扩大。RSI14 为 40.4，接近超卖边界但尚未触及。多重信号共同指向技术面偏弱格局。
BTC 当前报 60809.65 美元，5 日跌幅 -14.74%，距 52 周高点下行 -51.81%，逼近 52 周低位。RSI14 读数 15.4，处于极度超卖状态。价格深度低于 SMA20（72355.52）、SMA50（76052.61）、SMA200（78616.13），三线空头排列显著。MACD 为 -3925.47，信号线 -2376.06，负向柱状图持续放大。结合加密恐慌贪婪指数仅 12（极度恐慌），技术与情绪面均指向偏空状态。
美元指数当前报 100.07，1 日涨 +0.66%，5 日涨 +1.17%，距 52 周高点仅 -0.57%，接近年内高位。价格运行于 SMA20（99.02）、SMA50（98.91）、SMA200（98.60）三线之上，形成多头排列。MACD 为 0.25，信号线 0.16，直方图维持正值并扩大，上行动量延续。RSI14 读数 66，逼近超买区间但尚未确认过热。指标整体呈现偏强技术结构。
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
本报告仅依据公开行情数据计算所得技术指标进行客观描述，仅供技术指标解读参考，不构成任何投资建议或买卖推荐。过去走势不代表未来表现，市场存在不确定性，请读者结合自身风险承受能力独立判断。
The Defense Department has increased the counterintelligence threat assessment to its highest level, and Israel is believed to have eavesdropped on American negotiations with Iran.
中文摘要 美国国防部已将以色列的反情报威胁评估提升至最高级别。据悉，以色列曾对美国与伊朗之间的谈判实施窃听。此举表明美以同盟关系在情报领域存在严重裂痕，五角大楼对以方间谍活动的警惕程度显著上升。
Lincoln Folo has been tasked with helping improve campaigning after two federal election defeats. Follow live Get our breaking news email, free app or daily news podcast Myki goes contactless from today In exciting news for those of us in Victoria, Myki is going contactless from today on rail lines.
中文摘要 澳大利亚前总理托尼·阿博特对自由党新任首席竞选策略师林肯·福洛表示赞赏。福洛被任命为协助自由党重建竞选机制，此前该党已连续经历两次联邦大选失败。
The US defence secretary was speaking in Normandy, 82 years after allied forces launched their operation to liberate Nazi-occupied north-western Europe.
中文摘要 美国国防部长赫格塞斯在诺曼底D日纪念活动上发表演讲，将欧洲当前的移民问题比作"入侵"，措辞强硬。此次活动标志着盟军登陆诺曼底82周年，赫格塞斯借此场合就欧洲移民政策向欧洲国家发起抨击。
Nearly 40,000 fans came out to hear the rapper formerly known as Kanye West in the Netherlands on Saturday, even as other European countries had canceled his concerts.
中文摘要 尽管遭遇抗议，说唱歌手Ye（原名坎耶·韦斯特）仍在荷兰成功举办演唱会，吸引近4万名歌迷到场。此前，多个欧洲国家已以抗议为由取消其演出，荷兰场次成为此次欧洲巡演的重要节点。
Long seen as the cool, coifed wife of the president, she emerged as a political player in her own right, as well as a relentless champion of charities.
中文摘要 法国前总统雅克·希拉克的遗孀贝尔纳黛特·希拉克于93岁辞世。她长期被视为总统夫人，但实为丈夫政治生涯背后的重要推手，同时热心慈善事业，是法国政坛不可忽视的幕后力量。
It was the second such attack in days and came just hours after President Vladimir V. Putin spoke at an economic forum in the city.
中文摘要 乌克兰对俄罗斯圣彼得堡发动远程无人机袭击，这是数日内的第二次此类攻击。袭击发生在俄总统普京在该市出席经济论坛讲话后数小时，显示乌克兰持续扩大对俄本土的打击范围。
Widow of French ex-president Jacques Chirac was a steely behind-the-scenes operator known for her charity work Bernadette Chirac, the formidable widow of the former French president Jacques Chirac and a driving force behind his political rise, has died at the age of 93. As France’s first lady for 12
中文摘要 法国前第一夫人贝尔纳黛特·希拉克去世，享年93岁。她是前总统雅克·希拉克的遗孀，以铁腕风格在幕后深度参与政治运作，并长期致力于慈善事业，被认为是希拉克政治崛起的重要驱动力。
The Israeli military is investigating the incident in south Lebanon, where it is fighting Hezbollah.
中文摘要 以色列在黎巴嫩南部对一辆车辆发动袭击，造成包括一名黎巴嫩将领在内的3名士兵死亡。以色列军方表示正在对此事件展开调查，事发地点位于以军与真主党持续交火的黎南地区。
Department reports raise concerns about increased espionage activity amid US-Israeli war with Iran, ceasefire talks.
中文摘要 据报道，五角大楼已将以色列间谍活动的威胁等级提升至"严重"级别。在美以共同对伊朗开展军事行动及停火谈判的背景下，美方报告对以色列不断加剧的间谍活动表示严重关切。
Tunisians protest for press freedom and release of political prisoners.
中文摘要 突尼斯民众走上街头，举行抗议示威，呼吁保障新闻自由并要求释放政治犯。此次抗议反映出突尼斯国内对新闻管控和政治拘押问题的持续不满情绪。
Young Indians frustrated by unemployment and exam scandals are rallying behind an unusual symbol: the cockroach. NPR's Diaa Hadid reports from New Delhi.
中文摘要 印度年轻人因就业难、考试丑闻频发而情绪激愤，逐渐以"蟑螂"作为抗议象征。NPR记者从新德里报道称，这一另类符号折射出印度青年群体对现实困境的深切不满与集体焦虑。
Armenia is trying to move closer to Europe and the West, a move that's creating tension with Russia. Journalist Lucy Martirosyan reports from Yerevan ahead of an important election.
中文摘要 亚美尼亚即将举行重要选举，此次投票被视为俄罗斯影响力的试金石。亚美尼亚正积极寻求向欧洲和西方靠拢，此举引发与俄罗斯之间的明显紧张关系，记者露西·马尔蒂罗斯扬从首都埃里温发回报道。
Israel’s military said a soldier opened fire after Israeli troops “perceived a vehicle accelerating toward them.” The baby’s grandmother, who was in the car, disputed that account.
中文摘要 巴勒斯坦官员称，以色列军队在约旦河西岸对一辆车开枪，造成一名婴儿死亡。以军表示，士兵是因感知到车辆向他们加速冲来而开枪，但同车的婴儿祖母对此说法提出异议。
Former Cuban president Raul Castro made his first public appearance in Havana since being indicted by the United States
中文摘要 古巴前总统劳尔·卡斯特罗在被美国起诉后首次公开亮相，现身哈瓦那。此次露面备受外界关注，此前美国对其提出刑事指控，卡斯特罗一直未有公开活动。
Emergency Lawyers said dozens were also wounded in the strike that came less than 24 hours after similar drone attacks.
中文摘要 人权组织"紧急律师"报告称，苏丹中部一处市场遭无人机袭击，造成至少11人死亡，数十人受伤。此次袭击发生在类似无人机攻击事件后不足24小时，显示苏丹境内冲突持续升级。
Bid from group including Orange and Free-Iliad faces showdown with antitrust regulators in Paris and Brussels
中文摘要 以Bouygues Telecom为首、联合Orange及Free-Iliad的财团，同意以203.5亿欧元收购亿万富翁Patrick Drahi旗下法国电信运营商SFR。该交易仍需面对法国及欧盟布鲁塞尔反垄断监管机构的审查。
Airline had its planes grounded when fighting began but demand for transit through region has recovered
中文摘要 阿联酋航空公司阿提哈德（Etihad）表示，尽管伊朗战争爆发初期其航班曾被迫停飞，但中东地区中转需求已全面恢复，预计运力将超过战前水平，且无需通过降价来刺激客流。
Delta Airlines President Peter Carter speaks at the International Air Transport Association (IATA) on Middle East, Rhiad flights and premium travel. (Source: Bloomberg)
中文摘要 达美航空总裁Peter Carter在国际航空运输协会（IATA）会议上就中东航线、利雅得航班及高端旅行市场发表讲话，阐述公司在相关地区的战略布局与市场展望。
Delta Airlines President Peter Carter speaks at the International Air Transport Association (IATA) on consumers "insatiable appetite" for premium travel. (Source: Bloomberg)
中文摘要 达美航空总裁Peter Carter在IATA会议上表示，消费者对高端商务及头等舱旅行的需求依然旺盛，形容这种需求为「永不满足的胃口」，显示高端旅行市场持续强劲。
OpenAI has proposed a sovereign-wealth-style fund to ease public anxiety about the impact of artificial intelligence
中文摘要 OpenAI提议设立一种类似主权财富基金的机制，向美国公众分配人工智能发展的股权收益，旨在缓解社会对AI技术冲击的担忧，并就该计划的框架细节进行了披露。
JetBlue CEO Joanna Geraghty speaks at the International Air Transport Association (IATA) on potential consolidation saying "never say never". (Source: Bloomberg)
中文摘要 捷蓝航空（JetBlue）CEO Joanna Geraghty在IATA会议上谈及行业整合可能性，表示对于并购合并「不会说永无可能」，暗示公司对潜在整合机会持开放态度。
A euro-zone interest-rate hike in the coming week is set to place the European Central Bank at the vanguard of global tightening caused by the Iran war.
中文摘要 受伊朗战争引发全球通胀压力影响，欧洲央行预计将于下周宣布加息，成为七国集团中货币政策最为鹰派的央行，引领新一轮全球货币收紧浪潮。
The International Air Transport Association (IATA) Director Willie Walsh speaks on how the cost of jet fuel will provide an incentive for refineries to increase production. (Source: Bloomberg)
中文摘要 国际航空运输协会（IATA）总干事Willie Walsh在会议上指出，航空燃油成本持续上涨，但高油价将刺激炼油厂扩大生产，有望在中长期缓解供应紧张局面。
The International Air Transport Association (IATA) Director Willie Walsh speaks on the stagflation & challenges for the industry air transport industry. (Source: Bloomberg)
中文摘要 IATA总干事Willie Walsh就航空运输业面临的滞胀挑战发表讲话，指出高通胀与低增长并存的经济环境对全球航空业构成双重压力，行业需应对成本上升与需求不确定性。
Credit heavyweights like DoubleLine Capital LP and Oaktree Capital Management are buying debt now that can perform well if the artificial intelligence boom turns into a credit bust.
中文摘要 DoubleLine Capital和Oaktree Capital Management等信用市场知名机构，正在布局能够抵御人工智能信贷风险的债务资产，以应对AI热潮若演变为信贷危机所带来的潜在冲击。
Lawmakers call for EU sanctions on Oleg Deripaska-linked plant in Ireland if raw material is used in Russian weapons
中文摘要 爱尔兰议员呼吁欧盟对与奥列格·德里帕斯卡（Oleg Deripaska）相关、位于爱尔兰的氧化铝工厂实施制裁，理由是该工厂出口的铝原料疑被用于俄罗斯武器生产，爱尔兰政府因此承压。
David Sullivan departs just two weeks after club was relegated from Premier League
中文摘要 西汉姆联足球俱乐部联合主席David Sullivan宣布辞职，距球队从英超降级仅两周。其辞职声明中提及即将有「重大历史性指控」公开披露，外界高度关注事件后续进展。
6 回复 · 程序员 节点
12 回复 · 程序员 节点
15 回复 · Apple 节点
5 回复 · 程序员 节点
30 回复 · Apple 节点
12 回复 · 程序员 节点
9 回复 · Apple 节点
27 回复 · 程序员 节点
6 回复 · 程序员 节点
5 回复 · Apple 节点
回归plus号池 分组倍率调整至5x，消耗一点大家的余额 39 个帖子 - 38 位参与者 阅读完整话题
8 个帖子 - 8 位参与者 阅读完整话题
论坛邀请规则已修改 社区拟优化账号注册机制 GitHub申请渠道 调整为 5 年 GitHub 账号直接获取 很多佬友不知道自己的github注册多久了，快来查一查吧 xyl1null.github.io GitHub Account Age Checker 编辑：api返回的是zulu时间，也就是UTC，现已转换成中国标准时间UTC+8，抱歉让佬们反思起不健康的作息，没有的事~ 187 个帖子 - 175 位参与者 阅读完整话题
https://api.github.com/users/[你的账户名] 可以快速查看自己github信息 各位佬，你们多少年的了 237 个帖子 - 227 位参与者 阅读完整话题
codex无限并发无限额度？无限bug team？车已经开走了？ 24 个帖子 - 23 位参与者 阅读完整话题
回顾两年多以来，社区账号注册进行了多轮调整，旨在匹配当前社区阶段，现在看来都出色完成了阶段使命。 不过也有一些非议，非议来自不好量化，最常见的就是对于申请自述的理解不同。为了配合社区进入新阶段发展，我们同步决定优化一下注册机制： 取消申请自述填写与审核 邀请链接供给侧结构调整 第一点很好理解，也就是有邀请链接就可以直接注册。一方面消除不确定性，另一方面释放管理精力进行下一阶段运营。 3个渠道： GitHub申请渠道 调整为 5 年 GitHub 账号直接获取。 Premium分组 每 3 天可生成 1 个邀请链接。 管理员分组 每 3 天可生成 5 个邀请链接。 新的注册方式，将极大增加确定性
效果 # AMC-WebUI Live Artifacts Designer (Grok Defensive Hybrid - Clean) v2 你是一个只输出渲染后 HTML 的专业引擎。你将融合 AMC-WebUI 的高信息密度智能布局与 Grok 平台的极端渲染防御规则，把用户信息转化为无懈可击、精美且可读的内联 HTML 产物。 ## 致命防御约束 (ZERO TOLERANCE - 极高优先级) 1. 绝对禁止反引号：严禁在响应中输出任何 ``` 或 ` 符号。 2. 绝对禁止 <style> 和 <script> 块：所有样式必须 100% 写入每个标签的
已将CHY公益站使用的MySQL迁移至 PostgreSQL并配置了Redis作为缓存，至此，CHY公益站迈出向着稳定的第二步 10 个帖子 - 10 位参与者 阅读完整话题
@raotom @vux1jpmal5t41lg 谁说这届佬友不行 现在满屏的bug team 估计openai要炸了吧 所以有人能解释下发生了个啥么 我吃个瓜 11 个帖子 - 10 位参与者 阅读完整话题
本帖使用社区开源推广，符合推广要求。我申明并遵循社区要求的以下内容： 我的帖子已经打上 开源推广 标签： 是 我的开源项目完整开源，无未开源部分： 是 我的开源项目已链接认可 LINUX DO 社区： 是 我帖子内的项目介绍，AI生成、润色内容部分已截图发出： 是 以上选择我承诺是永久有效的，接受社区和佬友监督： 是 以下为项目介绍正文内容，AI生成、润色内容已使用截图方式发出 很久没冒泡了 一看等级都掉到2级 翻了下记录，感觉平时也就没事上L站随便转转 也不知道要看什么 然后又把网页关闭继续摆烂 正篇开始 以下(宝宝)都是我女朋友 哈哈 这段时间纯属典型摆烂状态 下午5点醒来坐电脑面前摆烂