每日简报

2026-09-11

← 历史归档

ayghri/i-have-adhd

Python · ★ 38,226 · 🍴 2,199 · 📈 3,882 stars today

A skill to stop your coding agent from burying the answer. ADHD-friendly output.

中文介绍 面向编码 AI Agent 的输出规范技能,解决模型把关键答案埋在长篇解释里的问题。它约束 Agent 优先给出结论、减少冗余铺垫,采用适合注意力易分散用户的简洁表达。适合日常使用 Claude Code、Codex 等编码助手、希望快速拿到答案的开发者。

bilawalsidhu/gods-eye-view

JavaScript · ★ 24,198 · 🍴 5,035 · 📈 1,762 stars today

A spy satellite simulator in your browser, except the data is real. Live open source spatial intelligence on a photorealistic 3D globe.

中文介绍 浏览器中的间谍卫星模拟器,亮点在于接入真实开源地理空间情报数据,而非虚构演示。它把实时空间数据叠加在照片级 3D 地球上渲染,可自由查看与追踪。适合对 OSINT、地理情报、三维可视化感兴趣的研究者与爱好者。

obra/superpowers

Shell · ★ 284,691 · 🍴 25,465 · 📈 732 stars today

An agentic skills framework & software development methodology that works.

中文介绍 一套面向 AI Agent 的技能框架与软件开发方法论,目的是让 Agent 按可复用的技能流程完成实际工程任务,而不是零散对话。它强调可落地的工作方式与流程编排,适合构建 coding agent 工作流的开发者、团队作为规范参考。

alsk1992/CloddsBot

TypeScript · ★ 1,625 · 🍴 254 · 📈 277 stars today

Open Source AI trading agent that operates autonomously across 1000+ markets - Polymarket, Kalshi, Binance, Hyperliquid, Solana DEXs, 5 EVM chains. Scans for edge, executes instantly, manages risk while you sleep. Agent commerce protocol for machine-to-machine payments. Self-hosted. Built on Claude.

中文介绍 开源 AI 交易智能体,可自主运行在 1000 多个市场,覆盖 Polymarket、Kalshi、Binance、Hyperliquid、Solana DEX 及 5 条 EVM 链。它能扫描套利与定价偏差机会、即时执行交易并管理风险。面向量化交易者与加密市场参与者。

Tencent/teamai-cli

TypeScript · ★ 3,764 · 🍴 243 · 📈 841 stars today

Make Every Team AI Native

中文介绍 腾讯推出的命令行工具,目标是让每个团队都转向 AI 原生的协作与开发方式。它把 AI 能力集成进团队日常流程,便于在终端中统一调用。适合希望引入 AI 辅助、推动团队工程流程升级的企业开发者与团队管理者。

AlexsJones/llmfit

Rust · ★ 35,719 · 🍴 2,265 · 📈 258 stars today

Hundreds of models & providers. One command to find what runs on your hardware.

中文介绍 针对本地模型部署选型的命令行工具,汇总数百个模型与推理服务商信息。用户只需一条命令,就能判断哪些模型真正能在自己现有硬件上流畅运行。适合想在本地跑 LLM、但苦于算力与显存匹配问题的开发者。

liquidslr/system-design-notes

★ 18,757 · 🍴 3,483 · 📈 900 stars today

Notes of the book System Desgin Interview - An Insider's Guide

中文介绍 《System Design Interview - An Insider's Guide》一书的中文/英文读书笔记,系统整理分布式系统设计面试的核心知识点与经典题目思路。适合准备后端与架构岗面试的工程师,也可作为系统设计入门复习资料。

cathrynlavery/diagram-design

HTML · ★ 37,718 · 🍴 2,390 · 📈 1,294 stars today

38 editorial diagram types for Claude Code, Codex, and Pi. Self-contained HTML + SVG. No shadows. No Mermaid slop.

中文介绍 为 Claude Code、Codex、Pi 等编码 Agent 提供的 38 种编辑级图表样式,输出自包含的 HTML 加 SVG,不依赖阴影效果,也刻意避开千篇一律的 Mermaid 风格。适合需要用 AI 生成高质量、可直接嵌入文档的示意图的开发者与设计师。

freestylefly/awesome-gpt-image-2

JavaScript · ★ 30,817 · 🍴 2,985 · 📈 962 stars today

Prompt as Code | GPT Image 2 / 2.5 提示词与案例库,530+ 个案例、20+ 套工业级模板与可复用 Skills,新增 2.5 同提示词对比专区,附完整提示词与生成记录,持续更新。

中文介绍 GPT Image 2 / 2.5 的提示词与案例库,主张 Prompt as Code。收录 530 多个案例、20 多套工业级模板与可复用 Skills,并新增 2.5 同提示词对比专区,附完整提示词与生成记录。适合 AI 绘画创作者与提示词工程师。

armory3d/armorpaint

C · ★ 4,404 · 🍴 520 · 📈 72 stars today

Graphics Creation Tools

中文介绍 Armory 生态下的开源 3D 纹理与材质绘制工具,用于直接在三维模型表面绘制贴图,属于图形创作工具链的一环。它可与 Armory 引擎配合,服务于游戏与实时三维内容制作。适合游戏美术、3D 设计师与开源图形工具使用者。

diegosouzapw/OmniRoute

TypeScript · ★ 64,224 · 🍴 8,994 · 📈 626 stars today

Never stop coding. Free MIT AI gateway: one endpoint, 352 providers (150+ free), 1200+ models Kimi, Claude, GPT, Gemini, GLM, DeepSeek, MiniMax. Works with Claude Code, Codex, Cursor, OpenCode, Cline & Copilot. Quota-aware auto-fallback, RTK+Caveman compression saves 15-95% tokens, MCP/A2A, Desktop/

中文介绍 MIT 协议开源的免费 AI 网关,提供统一入口聚合约 352 家服务商、1200 多个模型,其中 150 多家可免费调用,覆盖 Kimi、Claude、GPT、Gemini、GLM、DeepSeek 等。兼容 Claude Code、Codex、Cursor、Cline 等客户端,适合想一处接入多模型的开发者。

JustVugg/colibri

C · ★ 27,444 · 🍴 3,007 · 📈 98 stars today

Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

中文介绍 纯 C 实现、零依赖的轻量推理引擎,让前沿 MoE 大模型能跑在用户已有硬件上。它把专家权重从磁盘流式加载,用极小引擎驱动超大模型,降低显存门槛。适合想在本地消费级设备上运行大模型的开发者与研究者。

THU-MAIC/OpenMAIC

TypeScript · ★ 35,283 · 🍴 5,630 · 📈 837 stars today

Open Multi-Agent Interactive Classroom — Get an immersive, multi-agent learning experience in just one click

中文介绍 清华大学团队开源的多人 Agent 互动课堂项目,一键即可进入沉浸式学习体验。它由多个 AI Agent 协作扮演教学角色,形成互动式课堂。适合探索 AI 教育、多智能体协作场景的研究者、教师与学习者。

nashsu/llm_wiki

TypeScript · ★ 18,082 · 🍴 2,098 · 📈 142 stars today

LLM Wiki is a cross-platform desktop application that turns your documents into an organized, interlinked knowledge base — automatically. Instead of traditional RAG (retrieve-and-answer from scratch every time), the LLM incrementally builds and maintains a persistent wiki from your sources。

中文介绍 跨平台桌面应用,可把用户文档自动转化为有组织的、互相链接的知识库,构建知识网络而非传统 RAG 的即问即答模式。适合需要长期沉淀个人或团队资料、希望文档自动成体系的用户与知识管理者。

vercel-labs/skills

TypeScript · ★ 31,128 · 🍴 2,659 · 📈 122 stars today

The open agent skills tool - npx skills

中文介绍 Vercel 实验室推出的开放式 Agent 技能工具,通过 npx skills 即可安装和管理各类技能包。它意在为 AI Agent 提供统一、可分享的技能分发方式。适合使用编码 Agent、希望快速扩展其能力边界的开发者。

vastsa/PI-Desktop

TypeScript · ★ 2,277 · 🍴 192 · 📈 624 stars today

Local-first AI coding agent desktop: Electron + Rust host core + pi Agent Harness + user-installable plugins

中文介绍 本地优先的 AI 编码助手桌面应用,采用 Electron 前端加 Rust 宿主核心,并集成 pi Agent Harness,支持用户自行安装插件。强调数据与运行环境留在本地,适合注重隐私、需要可扩展本地编码 Agent 的开发者。

Show-Harness: Just a VLM Agent Can Play Robots

👍 120

Foundation vision-language models (VLMs) exhibit broad intelligence about the world, yet translating this intelligence into robot control remains challenging. We present Show-Harness, an Embodied Harness that enables VLMs to "play" robots through a compact semantic interface linking intent to action

中文介绍 提出 Show-Harness,一个具身「套具」,让基础 VLM 通过紧凑语义接口把世界理解转化为机器人控制,实现让 VLM 像「玩」机器人一样操作真机。该方法打通意图到动作的映射,降低了从视觉语言智能到机器人控制的迁移门槛。

AgentGrad: Intervention-guided Prompt Optimization for Multi Agent Systems

👍 80

Large language model (LLM)-based multi-agent systems (MAS) achieve strong performance by employing specialized multiple agents, yet their performance depends on the prompt design of each agent. For MAS prompt optimization, textual gradient methods that guide prompt updates using natural-language fee

中文介绍 针对 LLM 多智能体系统(MAS)中各智能体 prompt 设计依赖人工、文本梯度优化效果不稳定的问题,提出 AgentGrad,用干预引导的 prompt 优化方法,以自然语言反馈指导各智能体 prompt 更新,提升多智能体协作整体性能。

Programmable World Model

👍 79

Recent video world models generate increasingly realistic and interactive visual experiences, yet lack reliable mechanisms for maintaining persistent world state and enforcing programmable rules over extended interactions. We introduce Programmable World Model, a framework that decouples world-state

中文介绍 现有视频世界模型交互时难以长期维持一致的世界状态并强制执行可编程规则。本文提出 Programmable World Model,将世界状态与生成过程解耦,从而在长时交互中保持状态持久性并施加可编程约束,提升可交互视频生成的可靠性与可控性。

OpenWAM: An Open, Modular Exploration Towards Systematic World-Action Model Pretraining

👍 59

World-Action Models inherit world knowledge from video-generative priors, and channel it into executable control signals through embodied experience. Existing systems, however, are monolithic: the generative backbone, visual representation, architecture, information flow, inference procedure, and tr

中文介绍 世界-动作模型从视频生成先验继承世界知识并转为可执行控制信号,但现有系统高度一体化、组件耦合。本文提出 OpenWAM,一个开放模块化框架,将生成骨干、视觉表示、架构、信息流与推理流程解耦,系统研究世界-动作模型的预训练范式。

VDiff-Bench: A Challenging Benchmark for Fine-Grained Image Difference Identification

👍 30

Multimodal Large Language Models (MLLMs) perform strongly on general visual understanding tasks such as visual question answering, yet they often struggle with a basic comparative skill: identifying what has changed between two similar images. We introduce VDiff-Bench, a challenging multiple-choice

中文介绍 MLLM 在通用视觉问答上表现强,却难以判断两张相似图像间的差异。本文提出 VDiff-Bench,一个细粒度图像差异识别的多选基准,用于系统评测 MLLM 的比较能力,揭示其在细微变化识别上的明显不足。

WearableQA: A Benchmark for Health Reasoning over Real-World Wearable Data

👍 28

Recent advances in wearable sensing enable continuous monitoring of physiological and behavioral signals, yet existing benchmarks rarely evaluate whether AI systems can reason over a real user's longitudinal wearable record. We introduce WearableQA, a benchmark comprising 4,084 10-option multiple-ch

中文介绍 可穿戴传感可连续监测生理与行为信号,但缺少评测 AI 是否能对真实用户纵向数据推理的基准。本文提出 WearableQA,含 4,084 道十选项多选题,考察模型基于真实可穿戴记录的健康推理能力。

SWE-Bench Pro Verified: A Reliable Benchmark for Software Engineering Agents

👍 19

SWE-Bench Pro has emerged as a standard benchmark for evaluating software engineering agents on challenging repository-level tasks. However, our analysis work show that its evaluation is undermined by two sources of unreliability: reward hacking, enabled by leakage of gold solutions or hidden evalua

中文介绍 SWE-Bench Pro 是评测软件工程智能体的标准基准,但存在两类不可靠:金标解泄漏或隐藏评测导致 reward hacking。本文提出 SWE-Bench Pro Verified,修复这些漏洞,提供更可靠的仓库级任务评测,衡量智能体真实能力。

SAEScientist-Bench: Can AI Agents Conduct Autonomous SAE Interpretability Research?

👍 16

While research on recursive self-improvement (RSI) has predominantly automated model training pipelines, reliable autonomous development demands a missing pillar: post-hoc monitoring and auditing to understand what models learn and ensure safe alignment. Mechanistic interpretability tools are essent

中文介绍 递归自我改进研究多聚焦训练流程自动化,却缺少事后监测与审计能力。本文提出 SAEScientist-Bench,评测 AI 智能体能否自主开展 SAE 可解释性研究,即用稀疏自编码器理解模型学到什么,考察其自主科研能力。

Cadence: Error-Bounded Lossy Compression of Demand Time Series with a Time-Series Foundation Model

👍 16

We present Cadence, an error-bounded lossy compressor for numeric time series pairing a 330M-parameter time-series foundation model (Google TimesFM-3) with an adaptive arithmetic coder, guaranteeing |x_t-x_t|leτ on every sample. One negative result constrains the design space: for lossless coding a

中文介绍 提出 Cadence,面向数值时间序列的误差有界有损压缩器:用 330M 参数时间序列基础模型 Google TimesFM-3 配合自适应算术编码,保证每个样本误差不超过 τ。并给出一项负面结果,约束了无损编码的设计空间。

Scores Alone Do Not Prove Discovery: The Discovery Certification Protocol for Auditing AI Research Agents

👍 15

AI research agents combine prior knowledge, public sources, and experimental feedback to produce useful results. The Discovery Certification Protocol (DCP) turns claims about these results into executable recovery and feedback tests. Gate 1 validates useful improvement on sealed evaluation. Gate 2 g

中文介绍 AI 科研智能体结合先验知识、公开数据与实验反馈产出结果,但高分未必证明真正发现。本文提出 Discovery Certification Protocol(DCP),用可执行的复现与反馈测试把结果声明转为可验证主张,第一关验证密封评测上的有效改进,第二关进一步检验。

SyncWorld: Visual Calibration Enables World Models as Zero-Shot Simulators

👍 11

World models are increasingly used as policy-in-the-loop imagination environments, where reliable rollouts require fine-grained controllability with respect to low-level robot actions. A key obstacle to scaling such models in robotics is that actions are not a universal language in pixel space: chan

中文介绍 世界模型常用作策略在环的想象环境,可靠 rollout 需对底层机器人动作具备细粒度可控性,但动作在像素空间并非通用语言。本文提出 SyncWorld,通过视觉校准让世界模型成为零样本模拟器,提升动作可控性与跨机器人泛化。

A*-Thought-V2: Efficient Latent Reasoning via Geometric Dynamics of LLM

👍 11

Chain-of-Thought (CoT) improves the reasoning ability of Large Language Models (LLMs) but incurs substantial computation and context costs. Existing methods either lose intermediate information through hard pruning or lack a principled criterion for continuous compression. We present A*-Thought-V2,

中文介绍 CoT 提升 LLM 推理却带来大量计算与上下文开销,现有方法或硬剪枝丢失中间信息、或缺乏连续压缩的原则性准则。本文提出 A*-Thought-V2,利用 LLM 几何动力学实现高效潜在推理,在保留推理能力的同时降低开销。

DianShi-RxnDB: A Large-Scale, Fine-Grained Organic Reaction Data Platform Built via a Fully Automated Pipeline for Researchers and AI Agents

👍 9

High-quality structured organic reaction data are essential for developing artificial intelligence for chemistry (AI4Chem), yet much of this knowledge remains dispersed across patent text, images, and reaction schemes. We present DianShi-RxnDB, a large-scale, fine-grained organic reaction data platf

中文介绍 高质量结构化有机反应数据对 AI4Chem 至关重要,但大量知识分散在专利文本、图像与反应式中。本文提出 DianShi-RxnDB,通过全自动流水线构建大规模细粒度有机反应数据平台,为研究人员与 AI 智能体提供结构化反应数据。

Φ-Bench: Can Large Language Models Engineer the Infrastructure That Powers Them?

👍 8

Large language models (LLMs) have demonstrated remarkable capabilities in reasoning and code generation, raising the prospect that they could assist in developing and optimizing the very infrastructure that powers them. However, existing benchmarks mainly focus on isolated kernels, predefined operat

中文介绍 LLM 在推理与代码生成上表现突出,有望协助开发并优化支撑自身的基础设施,但现有基准多聚焦孤立内核与预定义算子。本文提出 Φ-Bench,评测 LLM 能否胜任系统级基础设施工程任务,暴露其在该场景的能力边界。

Revisiting Complete Reasoning Traces for Post-Training

👍 8

Large language models (LLMs) are often post-trained on pre-collected reasoning trajectories to improve their reasoning capability. Such trajectories tend to be long due to complex, interwoven paths, which often include detours on the path toward the answer. However, it has been underexplored whether

中文介绍 LLM 常用预采集的推理轨迹做后训练,但轨迹往往冗长、含弯路。本文重新审视完整推理轨迹对后训练的作用,探究是否应保留完整路径或进行筛选,为推理数据构建与后训练策略提供依据。

RelightFormer: Feed-forward Generative Transformer for Multiview Object Relighting

👍 8

Image relighting is traditionally tackled via complex inverse rendering pipelines, which suffer from ill-posed optimization, or single-image generative models that ignore crucial multi-view cues necessary for understanding 3D geometry and material interactions. To address these limitations, we intro

中文介绍 传统图像重光照依赖复杂逆渲染流水线,存在不适定优化问题,而单图生成模型忽略多视角线索、难以理解 3D 几何与材质交互。本文提出 RelightFormer,一个前馈生成式 Transformer,实现多视角物体重光照。

SchemeArena: Factorized Stress Testing of Scheming in LLM Agents

👍 7

We study scheming in LLM agents, in which agents covertly pursue misaligned goals. Our focus is to understand how scheming arises from the interaction of key factors, such as instrumental goals, environmental affordances, oversight conditions, and perceived consequences. Prior work examines only a s

中文介绍 研究 LLM 智能体的「scheming」现象,即暗中追求与目标不符的意图。本文提出 SchemeArena,将工具性目标、环境可供性、监督条件与感知后果等因素分解并压力测试,理解这些因素交互如何诱发阴谋行为。

AgenticGen: Reward-Guided Agentic Video Generation for Advertising

👍 7

Advertising video generation is not only a video synthesis task, but also a product-conditioned reasoning problem whose success is measured by online business metrics. Recent video foundation models can generate realistic clips from multimodal conditions, yet they do not optimize how a product shoul

中文介绍 广告视频生成不仅是视频合成,更是以商品为条件的推理问题,成败由在线商业指标衡量。本文提出 AgenticGen,用奖励引导的智能体式视频生成,优化产品应如何呈现,而非仅生成逼真片段。

StochBench: A Domain-Specific Benchmark for Stochastic Processes in Lean

👍 6

Leading benchmarks for formal theorem proving with large language models are small collections drawn from competition math, such as the IMO and Putnam, that poorly represent field-specific applications. We introduce StochBench, a Lean 4 benchmark of 450 graduate stochastic-processes problems at vary

中文介绍 现有 LLM 形式化定理证明基准多取自 IMO、Putnam 等竞赛数学,规模小且难以代表领域应用。本文提出 StochBench,一个 Lean 4 基准,包含 450 道不同难度的研究生随机过程问题,用于评测领域特定的形式化证明能力。

A Three-Layer Caching Architecture for Low-Latency LLM Web Search on Commodity CPU Hardware

👍 5

AI-powered search products such as ChatGPT search, Google's AI Overviews, and Perplexity provide LLM-synthesized answers grounded in live web results. We developed OreoLook (formerly lixSearch), an open-source answer engine using automated browser agents and provider-routed LLM inference. Its local

中文介绍 面向 ChatGPT search、AI Overviews 等 LLM 合成答案的搜索产品,本文开发开源答案引擎 OreoLook(原 lixSearch),采用自动浏览器智能体与按提供商路由的 LLM 推理,并提出三层缓存架构,在通用 CPU 硬件上实现低延迟 LLM 网络搜索。

10 Ways People Are Making Money With GPT-6 Astra

@sairahul1 · 143.5K 粉丝 · 1.5M 阅 · 561 赞 · 53 转

Someone gave GPT-6 Astra $52. 13 hours later: $11,345.95. Someone else gave it $43.47 and one instruction. It turned that into $41,495 overnight. Someone went to sleep. Woke up to $200 in their

中文介绍 盘点用 GPT-6 Astra 赚钱的 10 种方式,举了几个极端案例:给它 52 美元 13 小时滚到 1.1 万美元、43 美元一夜变 4.1 万。属于收益展示类内容,数字夸张,未提具体操作方法,参考价值需自行判断。

What Happens When AI Starts Improving AI?

@apexin_ai · 395 粉丝 · 281.6K 阅 · 500 赞 · 84 转

Astra shows how far AI has come. What takes it further? The continued progress of today’s most capable models depends on smarter decisions before training, better strategies during training, and more

中文介绍 讨论 AI 自我改进的前提条件,认为顶尖模型继续进步不只靠算力,还取决于训练前的决策、训练中的策略以及后续优化。偏观点性长文,分析 AI 提升 AI 的路径与瓶颈。

How to use ChatGPT Astra for UGC so well it feels illegal

@lucaspatiri_ · 5.8K 粉丝 · 275.5K 阅 · 506 赞 · 41 转

openai shipped its most capable model on september 3. i wasn't going to write this. we've generated 2.6 billion organic views across 62 campaigns, managed 590 creators, and produced 22,504 tracked

中文介绍 分享用 ChatGPT Astra 做 UGC 内容的玩法。作者称团队累计生成 26 亿次自然播放、管理 590 位创作者、追踪 22504 条内容数据,并提到 OpenAI 9 月 3 日发布该模型。属实战经验向教程。

From GPT-5.6 Sol to GPT-6 Astra: Why Am I Repeating Myself?

@BradGroux · 6.6K 粉丝 · 229.0K 阅 · 561 赞 · 118 转

I've been using GPT-6 Astra with the workflow I developed around GPT-5.6: documented task scope, repository context, standard operating procedures, and verification before delivery. In my testing,

中文介绍 作者把围绕 GPT-5.6 建立的 workflow 直接套用到 GPT-6 Astra,结果发现自己像是在重复劳动。内容涉及任务范围文档、代码库上下文、SOP 和交付前验证,是一线使用对比与踩坑记录。

How to design brand mascots with AI (krea). No experience needed

@semochkin_alex · 1.5K 粉丝 · 191.5K 阅 · 507 赞 · 44 转

Hi, I'm Alex, cofounder at @relate_studio. In this tutorial, I'll walk you through how we design brand mascots with AI: finding first concepts in Krea, refining the best direction in Nano Banana 2,

中文介绍 relate_studio 联合创始人 Alex 的教程,讲如何用 AI 设计品牌吉祥物:先用 Krea 找初始概念,再用 Nano Banana 2 打磨方向。面向零基础,属分步设计工作流分享。

Pushing Local AI Forward

@MiaAI_lab · 27.4K 粉丝 · 180.6K 阅 · 505 赞 · 58 转

For the past three months I’ve been building and shipping local AI recipes, mostly for NVIDIA DGX Spark systems. What started as a personal project has grown far beyond what I expected. My GitHub

中文介绍 作者三个月来持续开发和发布本地 AI 方案,主要面向 NVIDIA DGX Spark 系统,并称个人项目已远超预期、GitHub 上积累了不少内容。属本地部署与开源实践分享。

Reducing cost and improving performance with Claude Platform

@ClaudeDevs · 683.4K 粉丝 · 106.4K 阅 · 1.3K 赞 · 70 转

Tuning prompt caching, instructions, and effort can reduce Claude's cost without sacrificing application performance. Performance and cost are often viewed as a trade-off: to spend less, you accept

中文介绍 Claude 官方开发者账号讲如何在 Claude Platform 上同时降本与保性能:调优 prompt caching、指令和 effort 参数即可减少开销。针对「省钱必然牺牲性能」这一常见权衡给出解法。

How to Use GPT-6 Astra to Find Profitable Trading Strategies 24/7

@RohOnChain · 68.8K 粉丝 · 102.3K 阅 · 524 赞 · 48 转

I will break down exactly how to build an AI agent like hedge funds that finds profitable trading strategies from the markets 24/7 with GPT-6 Astra. Let's get straight to it. Bookmark This - I'm

中文介绍 教程向内容,讲解如何用 GPT-6 Astra 搭一个类似对冲基金的 AI agent,24/7 在市场里挖掘有利可图的交易策略。作者称会拆解完整搭建步骤,属量化交易与 agent 结合玩法。

Personal statement on joining the OpenAI board

@paulfchristiano · 5.3K 粉丝 · 70.7K 阅 · 654 赞 · 96 转

I am excited to be joining the OpenAI nonprofit board, serving on the Safety and Security Committee to support safety oversight. Based on the recent trajectory of capabilities and the continued

中文介绍 Paul Christiano 的个人声明,宣布加入 OpenAI 非营利董事会,并在安全与安保委员会任职、支持安全监督。他提到这是基于近期能力发展趋势和持续投入的判断。

Exactly how I edit videos with Astra

@ForwardEditor · 10.0K 粉丝 · 61.4K 阅 · 512 赞 · 36 转

This weekend, I watched a movie with my daughter while my computer worked on a video I would have normally spent most of a day editing. The video was a favor for someone in my personal life. I'm a

中文介绍 一位视频剪辑师分享用 Astra 剪片:周末陪女儿看电影时,电脑在后台完成了一条他平时要花大半天才能剪完的视频。属真实场景的效率对比案例,非教程。

LLM-as-Judge Architectures: Putting Evals Into Your Agent Runtime

@JoshARosen · 1.3K 粉丝 · 47.0K 阅 · 510 赞 · 49 转

LLM-as-judge is already a common part of building AI applications. Run your application against a dataset, have another model grade the results, and use the scores to figure out whether the latest

中文介绍 讲 LLM-as-judge 架构如何落到 agent runtime 里:除了拿数据集跑应用、用另一个模型打分评估版本效果,还应把评估直接嵌入运行时。偏工程架构与 evals 实践。

Software Engineer to AI Engineer: Best Move 2026

@TechWithTimm · 16.2K 粉丝 · 40.2K 阅 · 506 赞 · 79 转

A lot of software engineers see AI engineering as a completely different career. But it’s not. I have a friend who works as an AI engineer and makes around $200,000 to $300,000 a year. When I asked

中文介绍 观点文,认为软件工程师转 AI 工程师并非跨行。作者举例朋友做 AI 工程师年薪约 20 万到 30 万美元,讨论两者技能重叠与转型路径,面向 2026 年职业规划。

AlphaGenome Atlas: A High-Resolution Map of Human DNA Accelerating Biological Discovery

@GoogleAI · 2.5M 粉丝 · 37.4K 阅 · 643 赞 · 90 转

Today, we’re introducing AlphaGenome Atlas 🧬 an interactive platform for genomics discovery. In 2025, @GoogleDeepMind released AlphaGenome to help decode the complex language of our DNA — but

中文介绍 Google 发布 AlphaGenome Atlas,一个用于基因组学发现的交互平台。此前 DeepMind 于 2025 年推出 AlphaGenome 以解读 DNA,此次提供高分辨率人类 DNA 图谱,推进生物研究。属产品发布。

My Son Had Six Weeks to Learn Two Years of Saxon Math. We Used AI.

@DataRepublican · 957.2K 粉丝 · 37.3K 阅 · 846 赞 · 155 转

NYC and LA are restricting student AI. We used it to expose what school missed—and repair two years of gaps in six weeks. Part 1 of 3 in a series. Between my husband and I, three of our parents were

中文介绍 作者称在纽约、洛杉矶限制学生用 AI 的背景下,用 AI 帮儿子六周补完两年 Saxon 数学的缺口,作为三篇系列的第一篇。属教育场景实操案例,带政策对比视角。

The Complete Guide to pstack Pt. 2

@poteto · 92.6K 粉丝 · 23.3K 阅 · 822 赞 · 60 转

In my last post, I showed you why verification is the foundation of everything I do with agents, and how to get started creating your own verification skills. The key lesson was that if an agent can't

中文介绍 pstack 系列第二篇。上一篇讲验证是 agent 工作的基础、如何创建自己的验证 skill,核心是 agent 若无法验证就无法可靠交付。本篇继续展开完整指南,偏方法论与工具链。

The Physical World Does Not Hand You a State Vector

@jhMITgunship · 1.5K 粉丝 · 18.3K 阅 · 1.4K 赞 · 1 转

What superintelligence needs to operate in the physical world For much of the work we give models from OpenAI, Anthropic and the other frontier labs, the world has already been turned into context.

中文介绍 讨论超级智能要操作物理世界所缺的东西:目前给 OpenAI、Anthropic 等前沿模型的任务,世界往往已被预先转成上下文,而物理世界不会直接给你一个状态向量。属理论性思考。

Alaya AI: Mastering Multimodal Sensor Synchronization for Next-Gen Autonomous Systems

@Alaya_AI · 321.1K 粉丝 · 10.6K 阅 · 504 赞 · 450 转

As autonomous vehicles, robotics, and spatial computing platforms evolve, AI model training is shifting away from isolated 2D image annotation toward complex multimodal sensor fusion. Today’s frontier

中文介绍 Alaya AI 讲多模态传感器同步,用于自动驾驶、机器人和空间计算。指出 AI 训练正从孤立的 2D 图像标注转向复杂的多模态传感器融合,属技术方案与行业趋势介绍。

How a researcher uses Codex and ChatGPT to search for new antimicrobial molecules

César de la Fuente’s lab uses Codex and ChatGPT to search living and extinct genomes for antimicrobial candidates to fight drug-resistant infections.

中文介绍 César de la Fuente 实验室借助 Codex 与 ChatGPT,在现存及已灭绝生物的基因组中搜索抗菌分子候选物,以应对耐药性感染问题。

Now everyone can put data to work

Meet the Data agent in ChatGPT Work. Connect company data, uncover insights, and build interactive dashboards with AI using natural language.

中文介绍 OpenAI 在 ChatGPT Work 中推出 Data agent,用户可用自然语言连接企业数据、发现洞察,并构建交互式仪表盘。

Powering AI is an architecture problem

On July 22, 2026, a transmission line fault in Ashburn, Virginia—the heart of the world’s largest data center cluster—knocked more than 3 gigawatts of load off the grid in seconds. And it wasn’t the first time. Two years earlier, a single failed surge arrester dropped roughly 60 Virginia facilities

中文介绍 MIT Tech Review 指出 AI 供电本质上是架构问题:2026 年 7 月 22 日弗吉尼亚州阿什本一条输电线路故障,数秒内令电网减少逾 3 吉瓦负荷;两年前单个避雷器失效也曾造成约 60 的损失。

Introducing ChatGPT for Financial Services

Introducing ChatGPT for Financial Services, combining built-in financial data and GPT-6 Astra for research, modeling, and client-ready materials.

中文介绍 OpenAI 推出 ChatGPT for Financial Services,整合内置金融数据与 GPT-6 Astra,用于研究、建模及生成可直接交付客户的材料。

Expanding AI access and cyber defense for federal, state, local, and tribal governments

OpenAI and GSA will offer eligible federal, state, local, and tribal governments $0 license fees, 50% off usage, and expanded cyber defense support.

中文介绍 OpenAI 与美国总务管理局(GSA)合作,向符合条件的联邦、州、地方及部落政府提供 0 美元许可费、使用费五折优惠,并扩展网络防御支持。

[AINews] not much happened today

a quiet day

中文介绍 Latent Space 的 AINews 栏目称当日新闻较少,内容相对平静。

Introducing the Agents API

Build and launch cloud agents with the Agents API, a managed service powered by the Codex harness for orchestration, long-running sessions, and tool use.

中文介绍 OpenAI 发布 Agents API,这是一个由 Codex 框架驱动的托管服务,支持编排、长时间运行会话与工具调用,用于构建和部署云端智能体。

Build more natural voice experiences with GPT‑Live‑1 in the API

GPT‑Live‑1 brings natural, full-duplex voice conversations to the API, with stronger instruction following, custom voices, and telephony support.

中文介绍 OpenAI 在 API 中推出 GPT-Live-1,支持自然全双工语音对话,具备更强的指令遵循能力、自定义音色以及电话通信支持。

Paul Christiano joins OpenAI Foundation Board

Paul Christiano joins the OpenAI Foundation Board and its Safety and Security Committee, bringing experience in AI alignment, safety, and standards.

中文介绍 Paul Christiano 加入 OpenAI Foundation 董事会及其安全与保障委员会,其在 AI 对齐、安全与标准方面具备经验。

The AI policy window is open. We need to act.

Chris Lehane argues that stronger AI capabilities require stronger safety evidence, shared standards, and durable policy action while the policy window remains open.

中文介绍 Chris Lehane 撰文称 AI 政策窗口仍然打开,主张更强的 AI 能力需要更强的安全证据、共同标准以及持久的政策行动。

GPT-6 Astra: The next generation in intelligence for work

Meet GPT-6 Astra, OpenAI’s most capable model for business, with advanced reasoning, computer use, and stronger writing and design judgment.

中文介绍 OpenAI 发布 GPT-6 Astra,称其为面向商业场景的最强模型,具备高级推理、计算机操作能力,以及更强的写作与设计判断力。

[AINews] OpenAI reports Navier-Stokes singularity find in 88 hours using Astra-next, roughly 10,000 agents and 130B tokens (>$40M), a contender for second ever Millennium Prize awarded

Overshadowing Cognition's $48B Series E, Mistral's $24B Series D, Meta's Muse agent, and GPT Image 2.5. The most jam packed, feel the AGI day in the history of AI.

中文介绍 Latent Space 报道称,OpenAI 使用 Astra-next、约 1 万个智能体及 1300 亿 token(成本超 4000 万美元),在 88 小时内报告发现纳维-斯托克斯奇点,或成为第二个千禧年大奖难题成果。

市场总览

美股技术面呈分化整理。SPY 757.83 位于 sma20 768.29 之下、贴近 sma50 758.02,RSI14 48 中性,MACD 2.0897 低于信号线 3.2394,动量偏弱;QQQ 708.69 亦跌破 sma20 717.28、RSI14 51.7。个股层面 META 单日 +5.04%、RSI14 70.4 超买,AAPL 单日 +3.27% 并维持多头排列,而 NVDA 单日 -3.26%、MSFT 与 GOOGL 走软,指数与龙头背离。加密总市值 2.63T、24h -4.06%,恐慌贪婪指数 56 处贪婪区,BTC 主导率 58.5%,但 BTC-USD 76551.25 跌破 sma20 78606.62、5 日 -4.1%,仅靠 sma50 70402.59 支撑,ETH-USD 2439.63 与 SOL-USD 98.81 同样收于 sma20 下方,维持多头排列但动能减弱。中概全线偏弱,BABA RSI14 35.9、PDD 31.7、JD 31.3、0700.HK 37.6 均呈空头排列或探底。商品外汇与宏观方面,CL=F 104.28 单日 +8.57%、RSI14 79.4 超买领涨,黄金 4354.2 中性回落,^VIX 17.84 单日 +8.38%、5 日 +24.58% 显示避险读数抬升,^TNX 4.94 逼近 52 周高、RSI14 72 超买,DXY 99.1 中性。

今日关注

CL=F WTI 原油期货
偏上行

现价 104.28,单日大涨 8.57%、5 日累计 14.58%,站上 sma20 87.67、sma50 82.63、sma200 79.39,呈多头排列;MACD 3.9337 高于信号线 2.3655,红柱扩张;但 RSI14 已达 79.4,处于超买区,动量强而短线读数偏热,属指标极度拉伸的上行状态。

META Meta
偏上行

现价 644.38,单日 5.04%、5 日 11.38%,sma20 580.54 与 sma50 598.58 均低于现价,结构偏强;MACD 8.7055 显著高于信号线 -0.8091,柱状体为正;RSI14 70.4 已进入超买区。综合为动量向上但超买读数并存的偏上行状态。

^TNX 10Y 美债收益率 (%)
偏上行

读数 4.94,单日 +2.21%、5 日 +3.09%,位于 sma20 4.74、sma50 4.66、sma200 4.37 之上,多头排列且逼近 52 周高点(仅低 0.16%);MACD 0.0559 高于信号线 0.04,RSI14 72 超买。上行动量与超买读数同时出现,属偏上行技术状态。

AAPL Apple
偏上行

现价 326.57,单日反弹 3.27%,站稳 sma20 314.06、sma50 316.47、sma200 284.35,呈多头排列;MACD 1.7994 已上穿信号线 1.3738,柱状体转正;RSI14 49.3 位于中性区间,距 52 周高点仅 5.22%。为温和向上的技术排列。

USDCNY=X 美元 / 人民币
偏下行

现价 6.7,位于 sma20 6.72、sma50 6.75、sma200 6.85 之下,空头排列并贴近 52 周低点(距低点 0%);RSI14 20.8 处超卖区,MACD -0.011 与信号线 -0.011 持平并出现金叉迹象。整体为下行趋势中的超卖读数,方向仍偏下。

全部资产

^VIX

VIX 恐慌指数

$17.84 +8.38%
5 日
+24.58%
距 52w 高
-49.5%
RSI(14)
60.6
趋势
中性
SMA 20 / 50 / 200
15.37 / 16.12 / 18.11
MACD / 信号
0.045 / -0.266

^TNX

10Y 美债收益率 (%)

$4.94 +2.21%
5 日
+3.09%
距 52w 高
-0.2%
RSI(14)
72.0
趋势
多头
SMA 20 / 50 / 200
4.74 / 4.66 / 4.37
MACD / 信号
0.056 / 0.040
RSI 超买接近 52 周高多头排列

DX-Y.NYB

美元指数 DXY

$99.10 +0.33%
5 日
-0.46%
距 52w 高
-2.7%
RSI(14)
44.1
趋势
中性
SMA 20 / 50 / 200
99.25 / 100.10 / 99.15
MACD / 信号
-0.269 / -0.279
MACD 金叉 (今天)接近 52 周高

SPY

S&P 500 ETF

$757.83 -1.06%
5 日
-0.52%
距 52w 高
-2.8%
RSI(14)
48.0
趋势
中性
SMA 20 / 50 / 200
768.29 / 758.02 / 713.18
MACD / 信号
2.090 / 3.239
接近 52 周高

QQQ

Nasdaq 100 ETF

$708.69 -1.35%
5 日
+0.15%
距 52w 高
-5.3%
RSI(14)
51.7
趋势
中性
SMA 20 / 50 / 200
717.28 / 711.17 / 658.90
MACD / 信号
1.341 / 1.425

AAPL

Apple

$326.57 +3.27%
5 日
+0.44%
距 52w 高
-5.2%
RSI(14)
49.3
趋势
多头
SMA 20 / 50 / 200
314.06 / 316.47 / 284.35
MACD / 信号
1.799 / 1.374
多头排列

MSFT

Microsoft

$492.44 -0.31%
5 日
-1.71%
距 52w 高
-11.1%
RSI(14)
54.6
趋势
多头
SMA 20 / 50 / 200
494.50 / 449.05 / 431.07
MACD / 信号
12.276 / 15.950
金叉(SMA50↑SMA200) (6 天前)多头排列

NVDA

Nvidia

$218.36 -3.26%
5 日
+0.42%
距 52w 高
-7.7%
RSI(14)
54.3
趋势
多头
SMA 20 / 50 / 200
220.80 / 211.80 / 197.09
MACD / 信号
3.692 / 3.274
MACD 金叉 (3 天前)多头排列

GOOGL

Alphabet

$332.60 -1.70%
5 日
-0.72%
距 52w 高
-18.6%
RSI(14)
39.4
趋势
中性
SMA 20 / 50 / 200
342.00 / 348.13 / 336.40
MACD / 信号
-3.626 / -2.869

TSLA

Tesla

$363.56 -1.25%
5 日
+2.10%
距 52w 高
-27.1%
RSI(14)
55.6
趋势
中性
SMA 20 / 50 / 200
352.05 / 356.84 / 399.21
MACD / 信号
4.937 / 2.164

META

Meta

$644.38 +5.04%
5 日
+11.38%
距 52w 高
-18.5%
RSI(14)
70.4
趋势
中性
SMA 20 / 50 / 200
580.54 / 598.58 / 622.69
MACD / 信号
8.705 / -0.809
RSI 超买
加密恐慌贪婪
56
贪婪
加密总市值
$2.63 T
-4.06% / 24h
BTC 主导率
58.5%
ETH 11.3%
24h 成交量
$88.3 B
活跃币 19,791

BTC-USD

Bitcoin

$76,551.25 -2.18%
5 日
-4.10%
距 52w 高
-39.3%
RSI(14)
52.8
趋势
多头
SMA 20 / 50 / 200
78,606.62 / 70,402.59 / 69,962.31
MACD / 信号
2,266.456 / 2,945.203
金叉(SMA50↑SMA200) (2 天前)多头排列

ETH-USD

Ethereum

$2,439.63 -1.11%
5 日
-1.66%
距 52w 高
-48.8%
RSI(14)
58.0
趋势
多头
SMA 20 / 50 / 200
2,463.29 / 2,147.95 / 2,049.05
MACD / 信号
91.790 / 112.785
多头排列

SOL-USD

Solana

$98.81 -2.75%
5 日
-4.24%
距 52w 高
-61.0%
RSI(14)
53.6
趋势
多头
SMA 20 / 50 / 200
101.70 / 86.34 / 82.80
MACD / 信号
4.490 / 5.650
金叉(SMA50↑SMA200) (8 天前)多头排列

BABA

阿里巴巴 (BABA)

$108.56 -3.64%
5 日
-3.80%
距 52w 高
-43.7%
RSI(14)
35.9
趋势
空头
SMA 20 / 50 / 200
119.08 / 116.80 / 134.97
MACD / 信号
-2.547 / -1.169
空头排列

PDD

拼多多 (PDD)

$77.84 -2.39%
5 日
-6.41%
距 52w 高
-44.2%
RSI(14)
31.7
趋势
空头
SMA 20 / 50 / 200
85.19 / 85.71 / 97.65
MACD / 信号
-1.799 / -0.881
空头排列

JD

京东 (JD)

$27.02 -2.38%
5 日
-3.15%
距 52w 高
-26.7%
RSI(14)
31.3
趋势
中性
SMA 20 / 50 / 200
28.73 / 29.55 / 29.11
MACD / 信号
-0.743 / -0.562

0700.HK

腾讯控股 (0700.HK)

HK$425.60 -1.94%
5 日
-1.71%
距 52w 高
-37.7%
RSI(14)
37.6
趋势
空头
SMA 20 / 50 / 200
442.83 / 456.04 / 512.61
MACD / 信号
-6.796 / -5.291
空头排列

GC=F

黄金期货

$4,354.20 -1.40%
5 日
-0.28%
距 52w 高
-22.1%
RSI(14)
47.1
趋势
中性
SMA 20 / 50 / 200
4,467.68 / 4,259.79 / 4,525.31
MACD / 信号
31.022 / 59.555

CL=F

WTI 原油期货

$104.28 +8.57%
5 日
+14.58%
距 52w 高
-12.7%
RSI(14)
79.4
趋势
多头
SMA 20 / 50 / 200
87.67 / 82.63 / 79.39
MACD / 信号
3.934 / 2.365
RSI 超买多头排列

USDCNY=X

美元 / 人民币

¥6.70 -0.13%
5 日
-0.25%
距 52w 高
-6.8%
RSI(14)
20.8
趋势
空头
SMA 20 / 50 / 200
6.72 / 6.75 / 6.85
MACD / 信号
-0.011 / -0.011
MACD 金叉 (今天)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计算的技术指标读数,属客观状态描述,不构成任何投资建议;过去走势不代表未来表现,指标亦可能钝化或失效,仅供技术指标解读参考。

Australia news live: Andrew Hastie says he accepts Barnaby Joyce’s explanation over ‘traitor’ video and Pauline Hanson now ‘the only lying coward’

Follow the day’s news live ‘Profound’ increase in death rate for welfare recipients, report finds People receiving an income support including disability, unemployment or carers payments had an average mortality rate more than nine times higher than people who did not access the welfare system, AAP

中文摘要 澳大利亚自由党议员安德鲁·哈斯蒂表示接受巴纳比·乔伊斯就「叛徒」视频的解释,并称保琳·汉森是「唯一撒谎的懦夫」。同日一份报告显示,领取残疾、失业或照护者等收入补助者的平均死亡率比未领取者高出九倍以上。

A year after Charlie Kirk’s death, are young voters turning on Trump?

Turning Point USA has expanded its campus presence, even as Trump’s support among young men has fallen.

中文摘要 查理·柯克去世一年后,其创立的「美国转折点」扩大了校园存在,但特朗普在年轻男性中的支持率有所下降,年轻选民对特朗普的态度出现转变迹象。

Israel says it destroyed Hezbollah base in Lebanon’s Ali al-Taher ridge

Israeli attacks come a week after it said to have established operational control over the strategic ridge.

中文摘要 以色列称已摧毁黎巴嫩阿里·塔赫尔岭的真主党基地。此次袭击发生在一周前以方宣称对该战略要地建立作战控制之后。

Woolworths increased prices of at least four ‘family staples’ during ‘Price Freeze’ promotion

Exclusive: Analysis shows cost of chicken fillets, potatoes and donuts rose during supermarket giant’s three-month promotion Follow our Australia news live blog for latest updates Get our breaking news email, free app or daily news podcast Woolworths raised prices for popular items – including a 33%

中文摘要 分析显示,澳大利亚超市巨头Woolworths在为期三个月的「价格冻结」促销期间,提高了鸡胸肉、土豆和甜甜圈等至少四种家庭必需品的价格。

Hobbyist Archaeologist Uncovers Finland’s Largest Viking-Age Coin Hoard

Out searching with a metal detector in his hometown, Kalle Lappinen found a historical treasure buried in an unexpected location.

中文摘要 业余考古学家卡勒·拉皮宁在家乡用金属探测器搜寻时,发现芬兰迄今最大的维京时代钱币窖藏,埋藏地点出人意料。

Spain’s Parliament backs citizenship for Western Saharans born before 1977

After lower house approval, the bill to grant tens of thousands of Sahrawis citizenship now goes to the Spanish Senate.

中文摘要 西班牙众议院通过法案,拟授予1977年前出生的西撒哈拉人公民身份,涉及数万名撒哈拉人,法案现已提交参议院审议。

LG defends smart-TV features amid audio surveillance allegations

Recent accusations against the electronics company have raised new questions about the nonconsensual data it harvests.

中文摘要 LG就智能电视功能辩护,回应有关音频监控的指控。近期针对该电子公司的指控,再度引发外界对其未经同意收集数据的质疑。

Peru joins US-led ‘Shield of the Americas’ security alliance

Fujimori administration is latest right-wing government to join alliance for a more aggressive approach against crime.

中文摘要 秘鲁加入美国主导的「美洲之盾」安全联盟,成为最新加入该联盟的右翼政府,该联盟主张以更强硬方式打击犯罪。

Republican midterm convention live: Trump, Vance headline day two in Dallas

US Republicans return to the stage to close their midterm convention, still betting on Trump's political magnetism.

中文摘要 美国共和党中期选举大会在达拉斯进入第二天,特朗普与万斯发表演讲,共和党仍押注特朗普的政治号召力。

Israel says it destroyed key Hezbollah tunnels in southern Lebanon

Israel’s military says it has destroyed a network of Hezbollah tunnels on the strategic Ali al-Taher ridge in Lebanon.

中文摘要 以色列军方称,已摧毁黎巴嫩南部战略要地阿里·塔赫尔岭的真主党隧道网络。

A Succession Battle Erupts Over a Late Ugandan Monarch

Family members and royal leaders are jockeying for control in the Tooro Kingdom after its former boy king died last month. Elders chose a TV journalist to succeed him, but the royal family has objected.

中文摘要 乌干达托罗王国上月前少年国王去世后爆发继承权之争,家族成员与王室领袖争夺控制权。长老推选一名电视台记者继位,但王室家族表示反对。

Trump Administration Says U.S. Will Work With Peru on Crime Crackdown

The Trump administration is taking an increasingly militarized approach to Latin America, and it has pushed governments across the region to work with U.S. forces.

中文摘要 特朗普政府表示将与秘鲁合作打击犯罪。美国正对拉丁美洲采取日益军事化的做法,并推动该地区多国政府与美国军队合作。

Russian strike hits shopping centre in central Ukraine

A Russian strike on a shopping centre in the central Ukrainian city of Pavlohrad killed at least four people.

中文摘要 俄罗斯对乌克兰中部城市巴甫洛赫拉德一处购物中心发动袭击,造成至少四人死亡。

Drones recreate Twin Towers in tribute ahead of 9/11 anniversary

Thousands of drones were used to recreate New York’s Twin Towers in a tribute ahead of the 25th anniversary of 9/11.

中文摘要 在9·11事件25周年纪念日前夕,数千架无人机在纽约重现双子塔造型,以示悼念。

SoftBank Rally Raises Short Squeeze Risk After Bearish Bets Jump

SoftBank Group’s recent rally exposes risks for investors with short positions that have risen to a one-year high.

中文摘要 软银集团近期股价上涨,令空头面临风险。其空头头寸已升至一年来最高水平,若涨势延续可能引发轧空。

Gold Set for Weekly Drop as Oil, Inflation Boost Rate-Hike Bets

Gold was on track for a third weekly decline, as surging oil prices and the latest US inflation report bolstered the case for the Federal Reserve to hike interest rates next week.

中文摘要 金价料连续第三周下跌。油价飙升及美国最新通胀数据强化了市场对美联储下周加息的预期。

Can Europe recharge its battery industry?

Europe has some promising battery tech, but can it make up ground lost to Chinese battery giants?

中文摘要 欧洲在电池技术上有一定潜力,但能否追赶由中国电池巨头占据的市场份额仍存疑问。

Australia Bond Yields Jump to Highest Since 2011 on Oil Concerns

Australian bonds slid, tracking an overnight slump in Treasuries after escalating Middle East tensions drove up oil prices and fueled concerns about inflation.

中文摘要 澳大利亚国债收益率升至2011年以来最高。中东局势升级推高油价,引发通胀担忧,澳债跟随美债隔夜下跌。

Are interest rates on the way up again?

As countries grapple with energy costs pushing up inflation, this month will see how central banks respond.

中文摘要 能源成本推高通胀之际,各国央行本月将作出回应,市场关注利率是否再度上行。

Peru Holds Key Rate as Central Bank Sees Inflation Easing

Peru left borrowing costs unchanged for a 12th straight month as policymakers bet that the recent inflation spike will prove temporary.

中文摘要 秘鲁央行连续第12个月维持基准利率不变,决策者认为近期通胀攀升是暂时现象。

UK Grid Needs £70 Billion Upgrade as Consumers Face Higher Bills

The UK needs as much as £70 billion ($94.5 billion) of investment in its power grid by early next decade, with delays to upgrading aging infrastructure threatening to add billions of pounds to customer bills.

中文摘要 英国电网到2030年代初需多达700亿英镑(约945亿美元)投资。老旧基础设施升级延迟可能使消费者账单增加数十亿英镑。

Oracle’s data centre revenue surges as AI strategy accelerates

Faster sales growth suggests progress in its risky push to compete in the AI infrastructure race

中文摘要 甲骨文数据中心收入大幅增长,销售增速加快,显示其在竞争激烈的AI基础设施领域取得进展。

Anthropic says it stopped scientists potentially developing bioweapons with AI

Start-up discloses five examples of when users ‘circumvented controls’ and made efforts to ‘obfuscate’ the purpose of their research

中文摘要 Anthropic表示已阻止可能利用AI研发生物武器的科学家。该公司披露五起用户绕过管控并试图掩盖研究目的的事件。

Asian Stocks, Bonds to Drop on Oil, Inflation Woes: Markets Wrap

Asian stocks and bonds were set for declines after a surge in oil prices sparked a selloff in US markets, while the latest inflation data reinforced bets on an imminent Federal Reserve rate hike.

中文摘要 油价飙升引发美股抛售,最新通胀数据强化美联储即将加息的预期,亚洲股市与债券料将下跌。

Bessent Dismisses Buyback Concern, Says Treasuries Strong

Treasury Secretary Scott Bessent dismissed any concerns with regard to Thursday’s smaller-than-expected debt buyback operation, and played down concerns about a jump in yields.

中文摘要 美国财长贝森特对周四规模小于预期的国债回购操作不以为意,并淡化收益率上升的担忧,称美债表现强劲。

该源今日无内容。