每日简报

2026-08-16

← 历史归档

cordiverse/cordis

TypeScript · ★ 4,044 · 🍴 198 · 📈 616 stars today

Meta-Framework of Spatiotemporal Composability

中文介绍 一个面向时空组合性的元框架，为需要同时处理空间与时间维度的系统提供抽象层与编排机制，解决多源数据或组件在时空上的协同问题，适合地理信息、轨迹分析、仿真等场景。

cathrynlavery/diagram-design

HTML · ★ 18,571 · 🍴 1,114 · 📈 1,619 stars today

29 editorial diagram types for Claude Code. Self-contained HTML + SVG. No shadows, no Mermaid-slop.

中文介绍 专为 Claude Code 设计的 29 种编辑设计风格图表类型，采用自包含 HTML+SVG 实现，避免 Mermaid 的样式混乱和多余阴影，让 AI 直接生成清晰、专业、可嵌入的图表，适用于技术文档、汇报与数据可视化。

cursor/plugins

TypeScript · ★ 2,945 · 🍴 233 · 📈 152 stars today

Cursor plugin specification and official plugins

中文介绍 Cursor 编辑器的官方插件规范与插件集合，明确定义了插件开发标准，帮助开发者扩展 Cursor 的自定义工具、命令与工作流，适合需要为 AI 编码环境添加专属功能和集成第三方服务的团队。

cactus-compute/needle

Python · ★ 6,053 · 🍴 402 · 📈 551 stars today

14MB foundation model for tiny devices; phones, wearables, smart home, and robots.

中文介绍 面向手机、可穿戴设备、智能家居和机器人的微型基础模型，体积仅 14MB，可在资源受限的端侧设备本地运行，提供实时、低功耗的 AI 推理能力，适用于语音、感知和轻量任务处理。

unslothai/unsloth

Python · ★ 72,027 · 🍴 6,492 · 📈 435 stars today

Local UI to run and train LLMs and diffusion models, including Qwen3.8, Kimi K3, MiniMax-H3, Gemma 4, DeepSeek-V4, FLUX and more.

中文介绍 可在本地运行的图形界面工具，用于运行和微调 LLM 及扩散模型，支持 Qwen、Kimi、MiniMax、Gemma、DeepSeek、FLUX 等主流模型，无需云服务即可训练和推理，适合个人开发者低成本实验和部署。

public-apis/public-apis

Python · ★ 460,126 · 🍴 50,852 · 📈 2,476 stars today

A collective list of free APIs

中文介绍 聚合大量免费公开 API 并按类别整理的列表，涵盖天气、金融、新闻、游戏、社交等领域，帮助开发者快速查找和接入可用接口，是原型开发、功能验证和学习实践的常用资源库。

MakazhanAlpamys/Soup

Python · ★ 1,637 · 🍴 261 · 📈 303 stars today

Fine-tune LLMs from one YAML. Layer streaming trains an 8B model on a 4 GB laptop GPU.

中文介绍 通过单个 YAML 文件即可完成大模型微调，利用层流式（layer streaming）训练技术，让 8B 参数模型在 4GB 显存的笔记本 GPU 上也能训练，大幅降低微调门槛，适合资源有限的个人开发者和研究人员。

github/spec-kit

Python · ★ 129,174 · 🍴 11,548 · 📈 901 stars today

💫 Toolkit to help you get started with Spec-Driven Development

中文介绍 GitHub 出品的规范驱动开发（Spec-Driven Development）工具包，帮助团队以规格说明为核心组织编码、测试与评审流程，将需求直接转化为可执行规范，适合希望提升协作效率和交付质量的工程团队。

megadose/holehe

Python · ★ 13,108 · 🍴 1,742 · 📈 389 stars today

holehe allows you to check if the mail is used on different sites like twitter, instagram and will retrieve information on sites with the forgotten password function.

中文介绍 检测指定邮箱是否在 Twitter、Instagram 等网站注册的 OSINT 工具，通过调用各站点的“忘记密码”功能获取注册反馈，可用于安全研究、账号关联分析和隐私暴露排查。

altic-dev/FluidVoice

Swift · ★ 10,311 · 🍴 697 · 📈 165 stars today

Fastest and only macOS Dictation app with on-device STT and custom trained AI enhancement model. A local Wispr Flow alternative. DM us on X for an easter egg 😉 - https://x.com/fluidvoiceapp

中文介绍 macOS 上的本地听写应用，支持设备端语音转文字（STT），内置自训练 AI 增强模型，无需联网即可使用，是 Wispr Flow 的本地化替代方案，适合注重隐私的写作、笔记和语音控制场景。

ToolJet/ToolJet

JavaScript · ★ 39,517 · 🍴 5,293 · 📈 553 stars today

ToolJet is the open-source foundation of ToolJet AI - the enterprise app generation platform for building internal tools, dashboard, business applications, workflows and AI agents 🚀

中文介绍 开源的企业级应用生成平台，用于快速构建内部工具、数据看板、业务应用、自动化工作流和 AI Agent，提供可视化编辑器和丰富组件，帮助企业快速交付业务系统，也是 ToolJet AI 的开源基础。

HKUDS/CLI-Anything

Python · ★ 47,333 · 🍴 4,392 · 📈 100 stars today

"CLI-Anything: Making ALL Software Agent-Native" -- CLI-Hub: https://clianything.cc/

中文介绍 港大数据科学实验室推出的框架，致力于让所有软件实现“Agent 原生”，通过统一 CLI 接口和工具集，使 AI Agent 能直接调用各类应用，提升自动化操作能力，适合构建通用智能体和复杂工作流。

citrolabs/ego-lite

JavaScript · ★ 10,942 · 🍴 557 · 📈 546 stars today

The fastest browser for AI agents to run browser automation, built for sharing your logged-in browser state with your AI agents, like Codex or Claude Code, without disturbing you. Zero cost, zero config.

中文介绍 专为 AI Agent 浏览器自动化设计的轻量浏览器，能将当前登录状态安全共享给 Codex、Claude Code 等编程 Agent，零配置零成本，且不干扰用户操作，用于让 Agent 代替用户执行网页任务。

OpenART: Scaling Agent Red Teaming via Open-Ended Environment Evolution

👍 252

08/01 08:00

AI agents operate in persistent environments where early state changes can influence decisions far into the future. Unlike conventional language-model interactions, agent behavior is mediated through a shared state that is repeatedly modified and reused across long-horizon workflows. Current safety

中文介绍 针对AI智能体在持久环境中早期状态变化影响远期决策的安全测试难题，提出OpenART，通过开放式环境演化自动生成多样化挑战场景，扩展红队测试覆盖面，突破现有方法依赖固定场景的局限，提升对长时程任务中智能体风险的发现能力。

LLMRouter: Unified Infrastructure for Developing, Evaluating, and Deploying LLM Routers

👍 94

08/07 08:00

No single large language model (LLM) is optimal across all queries and budget constraints, making model routing essential for cost-effective deployment. Existing routers adopt diverse formulations and implementations, making fair comparison and extension difficult. We present a unified formulation o

中文介绍 为解决单一LLM无法在所有查询和预算约束下最优的问题，提出LLMRouter统一基础设施，对路由器开发、评估与部署提供统一表述，支持公平比较与便捷扩展，降低模型路由在成本效益部署中的使用门槛。

DreamX-Phi 1.0: Action-Conditioned Video World Model for Robotic Manipulation

👍 82

08/13 08:00

We present DreamX-Phi 1.0, an action-conditioned video world model for robotic manipulation that, given an observed frame, a language instruction, and a prescribed action sequence comprising end-effector poses and gripper states, predicts the resulting future observations. Yet realism alone does not

中文介绍 提出DreamX-Phi 1.0，面向机器人操作的动作条件视频世界模型，输入当前帧、语言指令和由末端执行器位姿及夹爪状态组成的动作序列，预测未来观测画面；并指出仅靠真实性不足，需进一步保证动作可控性与物理一致性。

Mechanist: AI as a Scientific Instrument for Discovering the Mechanisms of Intelligence

👍 82

08/12 08:00

AI models have achieved remarkable success across diverse domains, yet the mechanisms underlying their capabilities and the risks they may pose remain poorly understood. As AI development becomes faster and increasingly automated, mechanistic exploration remains largely manual, widening the gap betw

中文介绍 针对AI能力机制与风险认知滞后于模型发展的现状，提出Mechanist，将AI视为科学仪器，自动化地探索智能机制，旨在缩小自动化模型开发与人工机制分析之间的差距，提升对模型行为根源的理解。

SkillZip: Contract-Preserving Graph Compression for Scalable Agent Skill Libraries

👍 74

08/06 08:00

Large Language Models (LLMs) increasingly act as agents whose procedural knowledge is stored in reusable skill packages and loaded at inference time. As skill libraries grow, a central challenge is to expose the smallest sufficient executable context under a limited context budget. Existing systems

中文介绍 针对LLM智能体技能库增长带来的上下文预算限制问题，提出SkillZip，通过契约保持的图压缩，在有限上下文预算下提取最小且充分的技能执行上下文，降低技能加载成本，提升大规模技能库的可扩展性。

DarwinX: Evolving Agent Harnesses Through Natural Selection

👍 66

07/31 08:00

An LLM agent's capability depends not only on model weights but on its harness: prompts, tools, skills, and control flow. Self-improvement loops already edit harnesses, yet single-lineage search is path-dependent and local wins often regress other tasks. We introduce DarwinX, which treats self-evolu

中文介绍 针对LLM智能体自我改进中单谱系搜索路径依赖、局部提升造成其他任务退化的问题，提出DarwinX，将智能体harness（提示、工具、技能、控制流）的自我演化视为自然选择过程，通过种群化演化提升全局能力。

Intern-S2-Preview: Scientific Agentic Foundation Model

👍 43

08/13 08:00

Scientific discovery increasingly requires AI systems that can reason over scientific evidence of heterogeneous modalities, interact with scientific tools and environments, and sustain progress across long task horizons. We present Intern-S2-Preview, a series of scientific agentic foundation models

中文介绍 面向科学发现对多模态证据推理、工具交互和长任务持续推进的需求，提出Intern-S2-Preview系列科学智能体基础模型，能够处理异构模态科学数据并与科研工具环境交互，支持长时间尺度的科学推理任务。

How Can Rhetoric Reward-Hack AI Reviewers? Dissecting Rhetorical Sensitivity in AI-Based Peer Review

👍 39

08/10 08:00

As large language models increasingly participate in scientific evaluation, we investigate a potential form of reward hacking: how rhetorical choices shape AI-review judgments when reported scientific content is preserved and how these effects vary across evaluation conditions. We construct a contro

中文介绍 针对LLM参与科学评审可能被修辞策略奖励黑客的问题，通过构造控制实验，在保持科学内容一致的前提下，系统考察措辞选择对AI评审打分的影响及其在不同评估条件下的变化，揭示基于文本表述的评审漏洞。

AutoDesign: Meta-Harness Optimization for Long-Horizon Agentic Design

👍 35

08/13 08:00

Transforming multimodal sources into condensed and structured media outputs can be fundamentally conceptualized as a long-horizon agentic process centered on a model-harness system. While an ideal harness system should align with human design priors and accumulate reusable experience through empiric

中文介绍 将多模态源转化为结构化媒体输出视为围绕模型-harness系统的长时程智能体过程，提出AutoDesign，通过元harness优化使系统对齐人类设计先验并从经验中积累可复用知识，提升长期智能体设计的效率与质量。

PlayWorld: Benchmarking World Models with Agent Players over Long-Horizon Objectives

👍 35

08/13 08:00

Video world models simulate future states conditioned on current observations and user actions. Recent systems have demonstrated impressive video consistency and action controllability over long sequences. However, fairly comparing these interactive models remains challenging. In practice, a human p

中文介绍 针对视频世界模型在长序列一致性上表现优异但缺乏公平比较基准的问题，提出PlayWorld，引入智能体玩家在长时程目标下评估交互式世界模型，模拟人类使用方式，为不同模型的视频预测与动作控制能力提供统一评测框架。

Spatial Memory Agent: Experience-Grounded Procedure Memory for Spatial Intelligence

👍 30

08/13 08:00

Spatial intelligence is becoming a foundation for embodied agents, robotic planning, and multimodal assistants. To improve the spatial reasoning ability of VLM agents, existing work has mainly followed two lines. One line uses post-training methods, such as supervised fine-tuning and reinforcement l

中文介绍 针对VLM智能体空间推理能力不足的问题，提出Spatial Memory Agent，通过经验驱动的空间程序记忆机制，让智能体在交互中积累并复用空间操作经验，改善具身规划与多模态空间推理能力，避免依赖大规模后训练。

Can LLM Agents Stick to the Script? A Benchmark for Long-Horizon Consistency in Interactive Narratives

👍 26

08/08 08:00

The rapid advancement of Large Language Models (LLMs) is revolutionizing AI for Games by enabling open-ended and fluid interactive storytelling. However, existing research has largely overlooked the critical challenge of maintaining long-horizon logical consistency and narrative integrity against un

中文介绍 针对大语言模型在开放式交互叙事中难以维持长时程逻辑一致性与叙事完整性的问题，提出专项基准，测试智能体面对意外事件时能否坚守剧情设定，为AI游戏的叙事一致性评估提供标准化工具。

Massive Activations in Hybrid Linear Attention Large Language Models: Pre-Attention Spikes and Inter-Spike Plateaus

👍 19

08/12 08:00

We present the first systematic study of Massive activations (MAs) in layer-interleaved HLA LLMs and uncover two architecture-aligned morphologies: MAs consistently spike immediately before full attention layers, forming pre-attention spikes (PAS), and can persist through intervening linear attentio

中文介绍 首次系统研究层间混合线性注意力LLM中的大规模激活（MAs），发现其呈现两种与架构对齐的形态：在全注意力层前出现尖峰（PAS），并在相邻线性注意力层间形成平台，揭示大规模激活在混合架构中的分布规律。

Self-Evolving Embodied Agents via Skill-Harness Evolution

👍 11

08/11 08:00

Embodied agents are increasingly built as systems around foundation models, where performance depends not only on model weights but also on the skills, context, action interfaces, and execution harness surrounding the model. While supervised fine-tuning and reinforcement learning can adapt agents to

中文介绍 针对具身智能体性能受技能、上下文、动作接口与执行harness制约的问题，提出通过技能-harness演化的自我进化方法，使智能体在无需大规模微调或强化学习的情况下，自主改进外部组件来适应新任务。

Full-bandwidth transformer

👍 10

08/09 08:00

Autoregressive transformers compute along two axes: horizontally across generated tokens, and vertically through model depth. Dense attention gives each token broad horizontal access to the past, but the vertical feedback channel between decoding steps remains narrow: only the sampled token returns

中文介绍 针对自回归Transformer中解码步之间的垂直反馈通道狭窄、仅采样token回传的问题，提出全带宽Transformer，扩大模型深度方向的反馈带宽，让更多信息在解码步间流动，提升信息复用效率与生成质量。

Learning How the World Evolves: Extrapolative Video World Models via Latent Dynamics Reasoning

👍 10

08/10 08:00

The world evolves following its dynamics, i.e., its laws of motion. However, leading video diffusion models largely fit the pixels without modeling how the pixels transit over time. Thus, they render visually plausible frames but may not accurately obey the laws. To capture the dynamics purely from

中文介绍 针对视频扩散模型只拟合像素、不显式建模时间演化规律的问题，提出通过潜在动力学推理学习世界运动法则的外推式视频世界模型，从观测中提取潜在动态，生成符合物理规律的未来帧，提升外推能力。

An AI4AI Framework for Visual Token Pruning

👍 7

08/07 08:00

Visual-token pruning can substantially reduce the inference cost of multimodal large language models (MLLMs), yet existing methods largely rely on fixed, handcrafted heuristics and costly expert trial and error. As pruning objectives, budgets, and model architectures diversify, manually navigating t

中文介绍 针对多模态大模型视觉token剪枝依赖固定手工启发式、难以适应多样化目标与架构的问题，提出AI4AI框架，用AI自动搜索剪枝策略，在给定预算和模型条件下生成定制化剪枝方案，降低推理成本并减少人工试错。

The Illusion of Visual Tool-Use: A Causal Audit of Thinking with Images

👍 7

08/06 08:00

The "thinking-with-images" paradigm equips multimodal LLMs with active visual operations such as crop-and-zoom. However, models using these operations often achieve only marginal or negative gains over direct inference at substantially higher token cost. They may also repeatedly crop irrelevant regi

中文介绍 针对多模态LLM使用裁剪、缩放等主动视觉操作在同图推理时增益甚微甚至为负、且token成本高昂的问题，进行因果审计，揭示视觉工具使用的“幻觉”，表明模型常重复裁剪无关区域，质疑当前视觉推理范式的有效性。

Thought-Level Beam Search for Reasoning

👍 6

08/11 08:00

Test-time compute scaling is a primary driver of performance in large reasoning models (LRMs), but extreme inefficiency bounds current approaches, shifting the critical question from how much compute to spend, to where to allocate it. We formalize test-time reasoning as a constrained compute allocat

中文介绍 针对大型推理模型测试时计算极度低效的问题，将测试时推理形式化为受约束的计算分配任务，提出思维级束搜索，在给定预算下将计算优先分配到最有希望的推理路径，提升推理性能与计算效率。

AVA-Encoder: Towards Agent-Native Video Representation Learning

👍 6

08/12 08:00

Creative agents still lack an effective way to learn from high-quality human films, limiting their ability to produce cinematic-grade videos. A key challenge is the absence of a structured video representation that is both faithful to film content and directly usable for agentic reasoning and manipu

中文介绍 针对创意智能体无法有效从高质量人类影片中学习、缺乏既忠实于影片内容又可直接用于智能推理与操作的结构化视频表示的问题，提出AVA-Encoder，面向智能体的视频表示学习方法，使智能体更好地理解和生成电影级视频。

Cursor + SpaceXAI: the fastest iterating team wins

@sarahdingwang · 18.0K 粉丝 · 1.2M 阅 · 514 赞 · 73 转

08/14 22:28

By @BornsteinMatt, @martin_casado, and myself. AI coding could be the biggest addressable market of our lives. And, unless you think we’ve fully addressed it (which no one thinks; we are more like 1%

中文介绍 三位作者联合指出，AI 编码是此生最大的可寻址市场，目前开发程度仅约 1%；未来胜出的关键是迭代速度，Cursor 与 SpaceXAI 等工具正加速这一变革。

Agent Plugins are the future of Agent Skills

@GoogleCloudTech · 1.3M 粉丝 · 56.7K 阅 · 505 赞 · 76 转

08/13 10:49

Agent Plugins is an open, vendor-neutral standard for packaging Agent Skills and the MCP servers they depend on into one portable folder that any compatible client can load. Google is joining the

中文介绍 Google Cloud 推出 Agent Plugins 开放标准，将 Agent Skills 及其依赖的 MCP 服务器打包为可移植文件夹，任何兼容客户端都能直接加载，旨在推动 AI 技能生态的互操作性。

The Elon–Jensen "Understanding"

@TeslaLarry · 53.1K 粉丝 · 56.5K 阅 · 507 赞 · 54 转

08/14 00:20

After the SpaceX earnings call on August 4, several pieces started lining up for me in a way that is hard to ignore. Elon didn’t just praise Nvidia hardware, he declared an exclusive commitment to

中文介绍 作者从 8 月 4 日 SpaceX 财报电话会观察到，Elon 对 Nvidia 硬件不只是称赞，而是宣布独家承诺；这显示他与黄仁勋之间存在一种难以忽视的战略默契，或重塑 AI 算力格局。

Preparing GLM-5.3 for Open Release: A Responsible Path to Cyber Defense

@Zai_org · 136.2K 粉丝 · 42.7K 阅 · 583 赞 · 50 转

08/14 23:04

When GLM-5.2 helped Hugging Face investigate an incident in which an AI autonomously bypassed its own safeguards, it highlighted a broader shift. AI is becoming part of both cyber offense and cyber

中文介绍 智谱发布 GLM-5.3 开源前准备路径，聚焦网络防御责任。此前 GLM-5.2 曾协助 Hugging Face 调查 AI 自主绕过自身安全护栏的事件，凸显 AI 正同时进入网络攻防两端。

The AI Engineering Skills Map

@AndrewYNg · 1.8M 粉丝 · 41.2K 阅 · 716 赞 · 89 转

08/15 00:29

I am delighted to present The AI Engineering Skills Map. AI allows us to build software very differently today than in 2022, and everyone with the skills to take advantage of this shift has numerous

中文介绍 吴恩达发布 AI 工程技能图谱，指出 AI 已让软件开发方式与 2022 年截然不同；掌握这些新技能的人将拥有大量机会，图谱为开发者提供了清晰的能力进阶路径。

What's new in Gemini 3.7 Flash

@GoogleAIStudio · 192.9K 粉丝 · 31.5K 阅 · 535 赞 · 34 转

08/14 23:00

Gemini 3.7 Flash (gemini-3.7-flash) is generally available (GA) and ready for production use. It is our most intelligent workhorse model yet for software engineering, and agentic workflows. This guide

中文介绍 Google 宣布 Gemini 3.7 Flash 正式可用（GA），可直接用于生产环境；官方称其为面向软件工程与智能体工作流最智能的主力模型，并发布使用指南介绍新能力。

Introducing Gemini 3.7 Flash

@GoogleAIStudio · 192.9K 粉丝 · 30.7K 阅 · 566 赞 · 60 转

08/14 01:25

Today, we’re building on the progress of our widely used Flash series by introducing Gemini 3.7 Flash, our most intelligent workhorse model yet for coding and agents. This release comes just three

中文介绍 Google 推出 Gemini 3.7 Flash，距上一代仅三个月，定位编码与 AI 代理场景的最强主力模型，延续 Flash 系列兼顾性能与成本的特点，是 Gemini 3.x 系列的重要升级。

Qwen3.8-27B Practical Guide: Control Reasoning Depth and Extend Context to 1M Tokens

@QwenDevs · 8.0K 粉丝 · 25.6K 阅 · 559 赞 · 51 转

08/14 23:40

Qwen3.8-27B is here — a compact, deployment-friendly dense model built on the Qwen3.5 architecture, with native vision-language understanding and flexible thinking control. Two developer-facing

中文介绍 阿里发布 Qwen3.8-27B 实用指南，该模型基于 Qwen3.5 架构，紧凑易部署，原生支持视觉语言理解，并可通过控制思维深度将上下文扩展至 1M tokens，面向开发者提供双特性。

ORACLE: Confirmed AI Bots Actively Trading on Polymarket — Gains Reaching 200% Within 24 Hours

@OracleFoundtion · 19.4K 粉丝 · 18.8K 阅 · 3.1K 赞 · 615 转

08/15 05:48

In 2026, autonomous AI agents have become one of the most effective strategies on prediction markets. Over 30% of all activity on Polymarket now comes from algorithmic and AI-powered wallets. We

中文介绍 ORACLE 研究确认，2026 年自主 AI 代理已成为预测市场最有效策略之一；Polymarket 超 30% 交易量来自算法与 AI 钱包，部分代理 24 小时内收益高达 200%。

Flock is tightening its rules in response to a growing surveillance backlash

MIT Tech Review AI · 08/13 21:41

The police-tech giant Flock is announcing today that it will change officers’ access to its nationwide network of license plate readers, in an apparent effort to quell a growing backlash and win back contracts lost amid concerns about mass surveillance and police abuse. Several changes aim directly

The builder’s guide to GPT‑5.6

OpenAI News · 08/13 19:00

Learn how startups use GPT-5.6 to build faster, more cost-efficient AI agents with smarter model selection and new Responses API capabilities.

How kids feel about AI, in their own words

MIT Tech Review AI · 08/13 17:00

When we set out to talk to kids about artificial intelligence, we thought we knew what we’d hear. We expected some to tell us they were using it to cheat a little, the way Millennials and Gen Xers opened up CliffsNotes or programmed formulas into their TI-82s, and others to share inspiring ways they

not much happened today

Smol AI News · 08/13 13:44

**Google** rapidly released **Gemini 3.7 Flash** just three weeks after 3.6 Flash, targeting coding, web development, knowledge work, and agentic workflows with a 50% introductory price cut and improved benchmark scores like **DeepSWE 65.3%** and **Code Arena Elo 1588**. The update quickly integrate

市场总览

从四类资产整体看：美股主要指数处于上行趋势，SPY贴近52周高点（距高点仅-0.39%）且维持多头排列，QQQ同样多头排列，但内部出现分化——MSFT出现RSI超买（70.9），AAPL与GOOGL MACD均低于信号线、短线动量转弱；加密市场在恐慌贪婪指数34（恐慌）的背景下技术面整体偏弱，BTC-USD呈MACD死叉+空头排列，近5日-1.37%，ETH与SOL则围绕中性区间震荡，总市值2.25T、BTC主导率56.2%显示存量博弈；中概股普遍承压，JD近5日重挫-11.86%，0700.HK亦为死叉+空头排列，BABA、PDD虽未破位但MACD均处死叉状态；商品外汇方面，黄金动量强劲（MACD 68.6高于信号32.1），原油触发MACD金叉并站上所有均线，美元/人民币RSI 23.5深度超卖且贴近52周低点，下行压力显著。整体来看，风险资产技术面分化，原油与黄金偏强，加密与中概偏弱，美元资产处于超卖区域。

今日关注

CL=F WTI 原油期货 (CL=F)
偏上行

WTI原油当前价82.4，MACD（0.317）已上穿信号线（0.200）形成金叉，MACD柱由负转正，均线系统呈多头排列（SMA20 82.51＞SMA50 79.55＞SMA200 76.84），近5日累计上涨5.4%，RSI 52.5处于中性偏强区域，短线动量明显转暖，技术setup偏上行。

USDCNY=X 美元/人民币 (USDCNY=X)
偏下行

美元兑人民币当前价6.73，RSI 14日读数23.5已进入深度超卖区，MACD（-0.0101）低于信号线（-0.0084）维持死叉，价格位于SMA20（6.76）、SMA50（6.77）下方且呈空头排列，距52周低点幅度为0%（即处于区间低位），近5日下跌0.25%，下行趋势特征明确。

JD 京东 (JD)
偏下行

京东当前价29.06，近5日累计下跌11.86%，跌幅显著；RSI 37.8偏弱，MACD（0.447）低于信号线（0.910）为死叉状态，价格跌破SMA20（31.57）并呈空头排列，SMA200（29.38）与现价基本持平，说明中期支撑正在经受考验，技术面明显偏空。

META Meta (META)
中性

Meta当前价589.85，技术信号相互矛盾：MACD（-5.392）已上穿信号线（-5.604）构成金叉，但RSI 47.6处于中性区间，均线系统仍为空头排列（SMA20 594.67＜SMA50 597.13＜SMA200 627.63），距52周高点-25.41%，显示下跌趋势未扭转、短线动量刚出现修复迹象，方向尚不明确，定为中性。

SPY 标普500 ETF (SPY)
偏上行

标普500 ETF当前价776.34，距52周高点仅-0.39%，均线系统呈典型多头排列（SMA20 756.20＞SMA50 748.93＞SMA200 705.46），MACD（8.515）维持在信号线（6.257）上方，RSI 65.7接近但未进入超买区，近5日上涨0.4%，上行趋势保持完整。

全部资产

^VIX

VIX 恐慌指数

$14.25 -2.60%
5 日
-4.36%
距 52w 高
-59.6%
RSI(14)
40.1
趋势
空头
SMA 20 / 50 / 200
16.63 / 17.18 / 18.53
MACD / 信号
-0.759 / -0.485
空头排列

^TNX

10Y 美债收益率 (%)

$4.70 +1.19%
5 日
+0.77%
距 52w 高
-1.1%
RSI(14)
49.7
趋势
中性
SMA 20 / 50 / 200
— / — / —
MACD / 信号
— / —
接近 52 周高

DX-Y.NYB

美元指数 DXY

$99.64 -0.32%
5 日
+0.04%
距 52w 高
-2.1%
RSI(14)
39.1
趋势
中性
SMA 20 / 50 / 200
100.40 / 100.47 / 99.16
MACD / 信号
-0.248 / -0.159
接近 52 周高

SPY

S&P 500 ETF

$776.34 -0.20%
5 日
+0.40%
距 52w 高
-0.4%
RSI(14)
65.7
趋势
多头
SMA 20 / 50 / 200
756.20 / 748.93 / 705.46
MACD / 信号
8.515 / 6.257
接近 52 周高多头排列

QQQ

Nasdaq 100 ETF

$731.07 -0.14%
5 日
+1.11%
距 52w 高
-2.3%
RSI(14)
59.6
趋势
多头
SMA 20 / 50 / 200
704.13 / 712.95 / 650.62
MACD / 信号
5.524 / 1.144
接近 52 周高多头排列

AAPL

Apple

$305.93 +0.22%
5 日
-2.36%
距 52w 高
-11.2%
RSI(14)
43.7
趋势
中性
SMA 20 / 50 / 200
318.43 / 309.19 / 280.48
MACD / 信号
-2.461 / 0.187

MSFT

Microsoft

$495.40 -0.30%
5 日
-0.92%
距 52w 高
-10.5%
RSI(14)
70.9
趋势
中性
SMA 20 / 50 / 200
450.24 / 412.78 / 432.48
MACD / 信号
28.582 / 25.060
RSI 超买

NVDA

Nvidia

$225.16 -0.06%
5 日
+0.54%
距 52w 高
-4.8%
RSI(14)
63.0
趋势
多头
SMA 20 / 50 / 200
210.39 / 206.52 / 194.92
MACD / 信号
5.277 / 3.226
多头排列

GOOGL

Alphabet

$345.90 -0.13%
5 日
-2.37%
距 52w 高
-15.3%
RSI(14)
47.0
趋势
中性
SMA 20 / 50 / 200
346.41 / 353.89 / 331.38
MACD / 信号
-1.137 / -0.970
MACD 死叉 (今天)

TSLA

Tesla

$342.27 +0.68%
5 日
+4.17%
距 52w 高
-31.4%
RSI(14)
48.8
趋势
空头
SMA 20 / 50 / 200
329.14 / 371.09 / 406.08
MACD / 信号
-11.370 / -16.326
空头排列

META

Meta

$589.85 -0.86%
5 日
-0.38%
距 52w 高
-25.4%
RSI(14)
47.6
趋势
空头
SMA 20 / 50 / 200
594.67 / 597.13 / 627.63
MACD / 信号
-5.392 / -5.604
MACD 金叉 (今天)空头排列
加密恐慌贪婪
34
恐慌
加密总市值
$2.25 T
+0.05% / 24h
BTC 主导率
56.2%
ETH 10.1%
24h 成交量
$28.2 B
活跃币 18,423

BTC-USD

Bitcoin

$63,033.13 +0.09%
5 日
-1.37%
距 52w 高
-50.1%
RSI(14)
43.2
趋势
空头
SMA 20 / 50 / 200
63,828.54 / 63,511.11 / 69,371.23
MACD / 信号
-193.995 / -42.779
MACD 死叉 (4 天前)空头排列

ETH-USD

Ethereum

$1,881.27 +0.03%
5 日
+0.53%
距 52w 高
-62.0%
RSI(14)
51.1
趋势
中性
SMA 20 / 50 / 200
1,888.61 / 1,830.18 / 2,018.36
MACD / 信号
9.123 / 14.775

SOL-USD

Solana

$75.28 -0.06%
5 日
-0.88%
距 52w 高
-70.3%
RSI(14)
50.8
趋势
空头
SMA 20 / 50 / 200
74.40 / 75.97 / 82.03
MACD / 信号
0.111 / -0.109
空头排列

BABA

阿里巴巴 (BABA)

$123.81 +1.35%
5 日
-3.58%
距 52w 高
-35.7%
RSI(14)
55.0
趋势
中性
SMA 20 / 50 / 200
121.81 / 113.76 / 138.95
MACD / 信号
3.520 / 3.681
MACD 死叉 (今天)

PDD

拼多多 (PDD)

$84.79 +0.74%
5 日
-7.60%
距 52w 高
-39.2%
RSI(14)
43.9
趋势
中性
SMA 20 / 50 / 200
87.58 / 84.03 / 101.76
MACD / 信号
0.904 / 1.434
MACD 死叉 (1 天前)

JD

京东 (JD)

$29.06 -0.82%
5 日
-11.86%
距 52w 高
-21.2%
RSI(14)
37.8
趋势
空头
SMA 20 / 50 / 200
31.57 / 29.21 / 29.38
MACD / 信号
0.447 / 0.910
MACD 死叉 (2 天前)空头排列

0700.HK

腾讯控股 (0700.HK)

HK$440.00 -0.23%
5 日
-8.10%
距 52w 高
-35.6%
RSI(14)
40.4
趋势
空头
SMA 20 / 50 / 200
464.94 / 455.32 / 531.32
MACD / 信号
0.380 / 4.610
MACD 死叉 (2 天前)空头排列

GC=F

黄金期货

$4,437.30 +1.69%
5 日
+2.23%
距 52w 高
-20.6%
RSI(14)
64.6
趋势
中性
SMA 20 / 50 / 200
4,174.63 / 4,158.01 / 4,486.27
MACD / 信号
68.613 / 32.063

CL=F

WTI 原油期货

$82.40 +1.42%
5 日
+5.40%
距 52w 高
-31.0%
RSI(14)
52.5
趋势
多头
SMA 20 / 50 / 200
82.51 / 79.55 / 76.84
MACD / 信号
0.317 / 0.200
MACD 金叉 (2 天前)多头排列

USDCNY=X

美元 / 人民币

¥6.73 -0.16%
5 日
-0.25%
距 52w 高
-6.4%
RSI(14)
23.5
趋势
空头
SMA 20 / 50 / 200
6.76 / 6.77 / 6.89
MACD / 信号
-0.010 / -0.008
RSI 超卖接近 52 周低空头排列
风险提示

本报告仅基于公开行情价格计算的技术指标（均线、RSI、MACD等）进行客观状态描述，不构成任何投资建议或买卖指令。过去走势不代表未来表现，市场存在不确定性，技术指标亦存在滞后与失效可能。仅供技术指标解读参考，读者应结合自身情况独立判断并自担风险。

Bond Traders Agonize Over AI Companies’ $70 Billion of Shadow Credit Backstops

Bloomberg Markets · 08/16 03:00

Even before Nvidia Corp.’s splashy $500 billion financing partnership this week, investors were starting to fret over the roughly $70 billion in phantom liabilities that don’t appear on major AI companies’ balance sheets, but could materialize at the worst possible time.

中文摘要 债券交易员担忧AI公司约700亿美元的影子信用支持。这些表外负债未出现在主要AI公司资产负债表中，却可能在最不利时刻成为现实。此前英伟达刚宣布5000亿美元融资合作，令市场更加关注这类潜在风险。

Peru’s Economy Unexpectedly Slows as Fishing, Agriculture Slump

Bloomberg Markets · 08/15 23:40

Peru’s economy unexpectedly slowed in June due to a slump in fishing and agriculture output amid unusual weather conditions.

中文摘要 秘鲁6月经济意外放缓，主要受异常天气影响，渔业和农业产出大幅下滑，拖累整体经济活动。这一结果出乎市场预料，引发对经济前景的更多关注。

Pointed! Bloomberg's Weekly News Quiz For Risk-Takers

Bloomberg Markets · 08/15 23:37

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 Joe Mathieu, Carol Massar and Lisa Mateo as they play and check out the quiz for yourself at Bloomberg.com (Source: Bloomberg)

中文摘要 彭博推出策略性新闻竞猜游戏“Pointed”，在传统问答中融入策略元素，不仅测试玩家对新闻的了解，还考验其对答案的信心。主持人Joe Mathieu、Carol Massar和Lisa Mateo在节目中试玩，观众也可在Bloomberg.com体验。

Colombia Quake Tests New President Days Into Term

Bloomberg Markets · 08/15 23:20

Rescuers in western Colombia are still searching for survivors after a 7.4 magnitude earthquake killed at least 287 people. Bloomberg News Latin America Government Reporter Mie Dahl joins Bloomberg This Weekend from Bogotá and explains to hosts Joe Mathieu and Carol Massar that the disaster has also

中文摘要 哥伦比亚西部7.4级强震已致至少287人遇难，救援人员仍在搜寻幸存者。新任总统上任仅数日便遭遇重大灾害考验，彭博记者Mie Dahl从波哥大带来现场报道。

Drone Swarms Push AI Deeper Into Modern Warfare

Bloomberg Markets · 08/15 23:05

Bloomberg News Cyber Security & Tech Reporter Katrina Manson got a firsthand look at autonomous drone-swarm technology that allows multiple aircrafts to coordinate and strike targets even when communications or GPS are disrupted. Joining Bloomberg This Weekend, Mason explains to Carol Massar that th

中文摘要 彭博网络安全与科技记者Katrina Manson获得第一手观察，实地探访自主无人机蜂群技术。该技术能让多架无人机在通信或GPS中断时仍协同行动、打击目标，显示人工智能正更深地融入现代战争。

Lakers Deal Caps Billionaire Rush Into Sports

Bloomberg Markets · 08/15 22:03

Bob Iger and Josh Kushner moved from pursuing a Las Vegas NBA expansion team to agreeing within days to buy the Los Angeles Lakers in a record $12.5 billion deal. Bloomberg News Host of Sports City Randall Williams is on Bloomberg This Weekend with hosts Joe Mathieu and Carol Massar and explains tha

中文摘要 Bob Iger和Josh Kushner放弃申请拉斯维加斯NBA扩军球队，转而在数日内达成协议，以创纪录的125亿美元收购洛杉矶湖人队。这笔交易凸显亿万富翁抢购体育资产的势头。

Strong Earnings Keep PNC’s Agati Bullish on Stocks

Bloomberg Markets · 08/15 21:48

PNC's Chief Investment Officer Amanda Agati is on Bloomberg This Weekend with hosts Joe Mathieu and Carol Massar and says that strong earnings and economic fundamentals continue to support US stocks, even as elevated Treasury yields and midterm uncertainty threaten to bring more volatility. (Source:

中文摘要 PNC首席投资官Amanda Agati表示，强劲的企业盈利和经济基本面继续支撑美国股市，即便美国国债收益率上升及中期选举不确定性可能带来更多波动。她在彭博节目中阐述这一观点。

US-Iran War Risks Repeated Cycles of Fighting

Bloomberg Markets · 08/15 21:37

Shifting US objectives and neither side able to secure a decisive military victory make a lasting resolution difficult and Bloomberg Economics Defense Lead Becca Wasser says risks the the US-Iran conflict becoming a protracted war. Joining Joe Mathieu and Carol Massar on Bloomberg This Weekend, Wass

中文摘要 彭博经济国防分析师Becca Wasser表示，美国目标不断变化，且双方均无法取得决定性军事胜利，使美伊冲突难以持久解决，可能演变为反复循环的长期战争。她在彭博节目中发出这一警告。

By The Way: Headline You May Have Missed

Bloomberg Markets · 08/15 21:18

Mystery vacations, college campuses for sale, buying sorority unwritten rules, and half off sushi: Bloomberg News' Lisa Mateo joins Joe Mathieu and Carol Massar on Bloomberg This Weekend to share headlines you may have missed from over the week. (Source: Bloomberg)

中文摘要 彭博新闻的Lisa Mateo与主持人Joe Mathieu、Carol Massar一同回顾本周易被忽略的头条，包括神秘度假、大学校园出售、女生联谊会不成文规则、寿司半价等趣闻，为观众盘点轻松话题。

Trump's ‘Golden Age’ Pitch Meets Affordability Frustration

Bloomberg Markets · 08/15 21:00

President Donald Trump is touting a “golden age” for the US economy, but Bloomberg News White House Correspondent Catherine Lucey is on Bloomberg This Weekend and says persistent concerns over gas, food and housing costs are complicating that message ahead of the midterms. Lucey explains to hosts Jo

中文摘要 美国总统特朗普宣扬经济“黄金时代”，但彭博白宫记者Catherine Lucey指出，汽油、食品和住房价格持续高企引发的负担能力担忧，在中期选举前使这一宣传复杂化。

‘Outlier’ S&P 500 Earnings Strength Cheers Wall Street’s Bulls

Bloomberg Markets · 08/15 21:00

Corporate America is blowing past Wall Street’s expectations for a stellar earnings season, adding to the rosy backdrop for a stock market trading near a record high.

中文摘要 美国企业财报季表现远超华尔街预期，标普500指数成分股盈利强劲，令看涨人士振奋，为接近纪录高位的股市增添了乐观背景。

Robinson Says Activists Need Leverage to Win

Bloomberg Markets · 08/15 20:39

Author of the book 'From Presence to Power' Rashad Robinson is on Bloomberg This Weekend and explains to hosts David Gura and Christina Ruffini that visibility, media attention and public support do not amount to power unless activists build the leverage needed to change decisions and institutions.

中文摘要 《从存在到权力》作者Rashad Robinson在彭博节目中表示，可见度、媒体关注和公众支持并不等于权力，除非活动人士建立起能促使决策者改变的筹码。

该源今日无内容。