Python · ★ 32,136 · 🍴 2,670 · 📈 1,111 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 AI agent 技能，自动研究 Reddit、X、YouTube、Hacker News、Polymarket 等平台的热门话题，生成有据可依的综合摘要。适合需要快速获取近期舆论焦点或市场情绪的用户。
C++ · ★ 88,237 · 🍴 56,607 · 📈 65 stars today
Open Source Computer Vision Library
中文介绍 老牌开源计算机视觉库，提供超过2500种优化算法，涵盖图像处理、目标检测、人脸识别等。广泛应用于机器人、自动驾驶、安防监控等场景，支持C++/Python/Java等语言。
Shell · ★ 37,408 · 🍴 2,677 · 📈 1,103 stars today
Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop
中文介绍 为AI模型注入“品味”，避免生成平庸、模板化的内容。通过风格引导和约束机制，提升对话、文案、创意等输出的质量和独特性，适合内容创作者和AI应用开发者。
Python · ★ 186,441 · 🍴 32,077 · 📈 1,112 stars today
The agent that grows with you
中文介绍 自适应AI代理，能与用户共同成长。通过持续学习用户行为和偏好，在交互中优化自身能力，适合个人助理、教育陪伴等需要长期个性化服务的场景。
TypeScript · ★ 27,573 · 🍴 3,120 · 📈 554 stars today
An Open Source implementation of Notebook LM with more flexibility and features
中文介绍 Notebook LM 的开源替代品，提供更灵活的笔记与AI结合体验。支持多格式文档导入、智能摘要、问答等功能，适合研究者、学生和知识工作者构建个人知识库。
TypeScript · ★ 19,055 · 🍴 2,938 · 📈 183 stars today
Let's use AI to Earn!
中文介绍 利用AI工具赚钱的聚合平台，整合写作、设计、数据标注等任务，用户可通过完成AI相关任务获取收益。适合希望利用AI技能变现的自由职业者和兼职人员。
Rust · ★ 47,703 · 🍴 5,027 · 📈 322 stars today
an open source, extensible AI agent that goes beyond code suggestions - install, execute, edit, and test with any LLM
中文介绍 开源、可扩展的AI代理，超越代码补全：能安装、执行、编辑文件、运行测试，并兼容任意LLM。开发者可用它自动化开发工作流，如部署、调试和代码重构。
TypeScript · ★ 29,874 · 🍴 2,961 · 📈 309 stars today
Project N.O.M.A.D, is a self-contained, offline survival computer packed with critical tools, knowledge, and AI to keep you informed and empowered—anytime, anywhere.
中文介绍 自包含离线生存计算机，集成关键工具、知识和本地AI。无需网络即可提供导航、医疗指导、文件检索等功能，适用于灾难应急、野外探险或断网环境。
C++ · ★ 115,465 · 🍴 19,323 · 📈 158 stars today
LLM inference in C/C++
中文介绍 纯C/C++实现的LLM推理引擎，支持多种量化格式，可在CPU和GPU上高效运行。适合在笔记本、服务器甚至嵌入式设备上部署开源大语言模型，如Llama、Mistral等。
Python · ★ 7,589 · 🍴 732 · 📈 1,554 stars today
A vector index built on TurboQuant, written in Rust with Python bindings
中文介绍 基于TurboQuant的向量索引库，使用Rust编写并提供Python绑定。专为高吞吐量近似最近邻搜索设计，适合推荐系统、语义搜索等对性能要求高的场景。
Roff · ★ 72,705 · 🍴 16,277 · 📈 350 stars today
所有小初高、大学PDF教材。
中文介绍 中国大陆小学到高中及大学教材的PDF合集，覆盖各主要学科。为学生、教师和自学者提供方便的电子版教材获取途径，无需逐个寻找资源。
JavaScript · ★ 2,130 · 🍴 275 · 📈 262 stars today
OpenAI Plugins
中文介绍 OpenAI官方插件集合，为ChatGPT等产品扩展能力，如联网搜索、代码执行、第三方服务集成。开发者可参考此仓库了解插件架构并构建自己的插件。
TypeScript · ★ 13,083 · 🍴 920 · 📈 245 stars today
Desktop app to manage markdown knowledge bases
中文介绍 桌面端Markdown知识库管理工具，支持双向链接、标签、全文搜索。适合个人或团队用纯文本格式管理笔记、文档，并保持数据完全本地可控。
Python · ★ 13,844 · 🍴 1,848 · 📈 28 stars today
Useful tool to track location or mobile number
中文介绍 手机号位置追踪工具，通过公开的API或信号数据获取大致位置。需注意合法使用，主要用于教育研究或紧急情况下的粗略定位辅助。
Rust · ★ 1,547 · 🍴 36 · 📈 316 stars today
PostgreSQL in-database durable execution
中文介绍 PostgreSQL内嵌的持久化执行引擎，允许在数据库中运行可靠的任务流程。适合需要声明式工作流、事务保证和故障恢复的场景，如ETL、定时任务和事件驱动处理。
👍 73
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
中文介绍 针对代码语言模型适应软件演化的高成本与脆弱性问题，提出Code2LoRA，用超网络为每个仓库生成LoRA适配器，避免全量微调与长输入检索，实现低成本、可演化的仓库级适配。
👍 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
中文介绍 现有智能体仅响应显式用户请求，忽视隐含问题。TIDE通过模板引导的迭代探索主动发现多个问题，在文档、工具和代码场景中提升问题覆盖率。
👍 38
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
中文介绍 现实规划需同时满足世界约束与用户约束，且约束逐步揭示。AdaPlanBench评估LLM智能体在渐进披露双重约束下的自适应规划能力，填补现有基准空白。
👍 31
We introduce MMAE, a Massive Multitask Audio Editing benchmark, serving as the first comprehensive evaluation testbed designed for general-purpose instruction-based audio editing. Spurred by the shift toward intelligent creation, interactive editing has rapidly expanded from visual domains, pioneere
中文介绍 MMAE是首个面向通用指令音频编辑的大规模多任务基准，涵盖多种编辑操作与音频类型，为评估交互式音频编辑提供系统测试平台。
👍 31
Large language models exhibit impressive zero-shot capabilities across a wide range of downstream tasks. However, they struggle to function as off-the-shelf embedding models, leading to suboptimal performance on massive text embedding benchmarks. In this paper, we identify a potential cause underlyi
中文介绍 发现LLM的解嵌入矩阵可视为文本嵌入的特征透镜，通过简单变换将其转化为有效嵌入模型，显著提升零样本文本嵌入性能。
👍 25
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,
中文介绍 家庭机器人常面临价值冲突场景（如效率 vs 自主性）。RobotValues评估机器人如何在任务完成之外平衡人类价值观，提供冲突情境下的行为准则。
👍 20
Despite being a pivotal frontier, interactive world modeling remains underexplored in terms of the versatile controllability required by practical scenarios. To bridge this gap, we present AnchorWorld, a framework that advances egocentric simulation through enhanced interaction integrity and a flexi
中文介绍 AnchorWorld提出基于视图演化定制的具身第一人称世界模拟框架，增强交互完整性与灵活性，支持用户自定义模拟演化过程。
👍 16
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
中文介绍 测试视频生成模型输出的物理可行性：将生成视频直接转换为机器人可执行动作，评估模型对物理世界的理解程度。
👍 14
Persistent AI assistants, such as OpenClaw, accumulate large collections of related memories over long-term interactions. As these memories grow, they may reinforce one another, diverge across contexts, or directly conflict, making correct assistance depend on memory relations rather than isolated r
中文介绍 持久AI助手需管理大量相关记忆。SubtleMemory基准测试智能体对记忆间细粒度关系（强化、分歧、冲突）的判别能力，超越孤立检索。
👍 13
Existing benchmarks evaluate Tool-Integrated Reasoning (TIR) in LLMs on idealized ''happy paths'', largely overlooking real-world tool failures. We introduce ToolMaze, a benchmark for dynamic path discovery and error recovery in TIR agents. To separate systematic replanning from blind trial-and-erro
中文介绍 现有工具集成推理基准仅考虑理想路径。ToolMaze引入动态路径发现与异常恢复评估，通过迷宫任务区分系统性重规划与盲目试错。
👍 12
Video understanding is being rapidly transformed by multimodal large language models (MLLMs), as research moves from short clips to long, multimodal, and knowledge-intensive video scenarios. These scenarios require models to handle sparse evidence, long-range dependencies, multimodal alignment, and
中文介绍 多模态大模型（MLLM）视频理解从短片段扩展到长视频、多模态知识密集型场景。提出「观察-记忆-推理」框架，处理稀疏证据与长程依赖。
👍 10
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
中文介绍 VLA模型借助VLM世界知识实现指令跟随操作，但语义空间与控制策略存在结构不匹配。AffordanceVLA通过可交互感知理解生成精确动作，弥合语义-动作鸿沟。
👍 9
While Vision-Language Models (VLMs) have shown strong visual reasoning capabilities, their spatial reasoning abilities remain largely constrained to the observed images and text-oriented chain-of-thought. They often struggle to infer unobserved layouts, maintain cross-view consistency, and reason fr
中文介绍 VLM空间推理局限于观察图像与文本思维链。提出「用想象思考」，结合世界模拟器进行具象空间推理，实现未观察布局推断与跨视角一致性推理。
👍 9
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
中文介绍 交互式图像编辑需要统一的对象与材质选择。MAOAM利用VLM同时支持文本/点击指定，实现对象与材质（如金属、布料）的灵活选择。
👍 8
We present dots.tts, a 2B-parameter continuous autoregressive text-to-speech (TTS) foundation model that models speech in a continuous latent space. Compared with existing continuous autoregressive models, our key innovations are threefold. First, we train an AudioVAE with multiple objectives to bui
中文介绍 dots.tts是一个2B参数连续自回归文本转语音基座模型，在连续潜空间建模语音。创新包括多目标AudioVAE与复合训练策略，提升生成质量。
👍 8
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
中文介绍 现有记忆化评估大多测量模型在强制提示下的数据泄露，而非普通使用。PropMe提出倾向感知框架，基于前缀与无前缀生成对比评估真实泄露倾向。
👍 7
Causal graphs provide a high-level language for making mechanisms transparent. Recent work uses Large Language Models (LLMs) to recover causal graphs of external-world processes. Instead, in this paper, we use causal graphs to model LLM inference itself, providing stakeholders with a transparent vie
中文介绍 利用因果图建模LLM推理过程本身，为利益相关者提供透明解释。通过反事实链追溯推理路径，揭示模型决策中的因果关系。
👍 7
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
中文介绍 时序定位通常只检索单片段。提出「一对多时序定位」，针对单查询定位多个不相交片段，并构建对应基准与方法。
👍 5
Self-evolving agents requires adaptation after deployment, but existing approaches assume a usable learning loop, such as curated skills, successful trajectories, or verifier signals. Real open-world deployments may provide none of these, offering only a task prompt. In this work, we study open-worl
中文介绍 自进化智能体需在开放世界部署后适应，但可能缺乏精心设计的技能/轨迹/验证信号。OpenSkill仅靠任务提示实现开放世界自我进化，无需外部监督。
👍 5
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
中文介绍 视频MLLM对每帧独立编码，造成大量重复视觉token。AdaCodec提出预测性视觉压缩编码，利用帧间冗余减少token数量，提升视频处理效率。
@sairahul1 · 111.8K 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 介绍2026年成为AI工程师的方法，无需CS学位或集训营，指出当下公司招聘并不看重Transformer等基础理论知识，而是更重视实践能力。
@sairahul1 · 111.8K 粉丝 · 546.4K 阅 · 536 赞 · 94 转
In February 2026, a small OpenAI team shipped 1 million lines of production code. They didn't write a single line by hand. The AI agents wrote it. The humans designed the system that made the agents
中文介绍 通过OpenAI团队用AI agent自动生成100万行生产代码的案例，揭示2026年AI工程师的核心技能转向设计agent工作系统（Harness Engineering），而非手写代码。
@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 公开一套完全自动化的YouTube增长系统，无需制作或上传任何视频，已获得12.5万订阅者，并声称可在3小时内获取10万订阅者。
@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 指出进入前沿AI实验室主要靠两件事：经得起验证的研究能力和一线工程能力，且二者正日益融合，不再是分离的技能。
@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 将广告之父David Ogilvy的写作规则（曾为客户创收超8.64亿美元）输入Claude，打造出一款传奇AI写作教练，分享其构建过程。
@rasbt · 459.9K 粉丝 · 50.9K 阅 · 507 赞 · 51 转
Catching up with the agent-related research literature, one paper that definitely got my attention is "Evaluating AGENTS.md: Are Repository-Level Context Files Helpful for Coding Agents?." It looks
中文介绍 推荐一篇论文，评估仓库级上下文文件AGENTS.md对编码agent的实际帮助，探讨此类文件是否真的能提升agent性能。
@ByteMohit · 2.0K 粉丝 · 43.3K 阅 · 501 赞 · 51 转
Everyone is building with agents. Almost nobody talks about what is actually inside one. Not the model. The harness around it. I spent the last few months building one from scratch in Python every
中文介绍 从零用Python构建Agentic Harness，揭示Agent的本质并非模型本身，而是围绕模型的围栏系统，分享数月构建中的见解。
@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 观察到多位PhD获得者春招立即获高薪岗位后，探讨直接进入产业的可能性，并整理2026年强化学习（RL）面试题。
中文介绍 Hugging Face博客报道了一个「神奇数字假牙」项目，最终以失败告终。
a quiet day of RSI.
中文介绍 Latent Space的AI新闻简报表示当日市场平静，仅提及RSI（相对强弱指标）相关情况。
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 文章讨论了如何避免发布低质量强化学习环境，指出有问题的环境会使模型变差，并提供了修复示例。
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 2026年6月5日，404 Media报道称攻击者利用Meta的AI客服代理窃取Instagram账户，只需请求将账户链接至攻击者控制的邮箱，AI即执行。事件凸显AI安全不止于Mythos。
a quiet day
中文介绍 Latent Space当日AI新闻简报表示市场平淡，无重大事件。
**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系列引发不同反应：Claude Mythos一次性工作流受好评，但Opus 4.8基准测试出现退化；Opus 4.7在化学任务表现强劲；Sakana AI推出了RSI相关项目。
中文介绍 当日AI新闻摘要：Anthropic Oceanus模型泄露、ChatGPT推出「Dreaming」记忆系统、递归自我改进技术进展。
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与Andon Labs的Lukas Petersson和Axel Backlund对话，讨论VendingBench评估体系，涵盖从Haiku到Mythos的Claude模型评估，以及如何从头构建前沿评估。
中文介绍 NVIDIA发布Nemotron 3.5内容安全模型，提供可定制的多模态安全方案，适用于全球企业AI。
中文介绍 ServiceNow AI发布EVA-Bench Data 2.0，覆盖3个领域、121种工具和213个场景，用于评估AI代理。
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 Endava利用AI代理、ChatGPT Enterprise和Codex重新设计软件交付流程，加速开发、自动化工作流，构建企业AI原生文化。
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 美国科罗拉多州联邦治安法官Maritza Braswell每天处理大量由AI生成的诉讼文件，这些文件多由无律师的人提交，法院正在应对AI带来的案件激增。
ChatGPT introduces a new memory system to better remember preferences, keeping context fresh and relevant across conversations.
中文介绍 OpenAI推出ChatGPT「Dreaming」记忆系统，用于更好地记住用户偏好，在对话中保持上下文新鲜和相关。
**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发布完全开源的Nemotron 3 Ultra模型，550B参数MoE架构，55B活跃参数，支持1M上下文，针对长时代理任务优化，速度提升5倍、成本降低30%，采用混合Mamba/注意力机制和LatentMoE。
a quiet day.
中文介绍 Latent Space报道当日AI新闻：Reve 2和Ideogram 4在图像生成中引入布局功能，整体市场平静。
当前市场技术面整体显著偏空。加密资产方面，恐慌贪婪指数仅8（极度恐慌），总市值2.25T美元（24h +1.28%），BTC主导率56.1%、ETH占比8.9%；BTC、ETH、SOL均处RSI超卖区（分别26.0、26.8、26.8），价格远低于所有主要均线，空头排列明确。美股方面，SPY跌2.58%至737.55，低于20日均线但仍守于50/200日均线之上，RSI 49.4中性；QQQ跌4.8%，MACD死叉；多数个股如NVDA、TSLA、META均跌破短期均线，RSI回落至40-44区间，趋势偏弱。VIX恐慌指数暴涨39.68%至21.51，突破所有均线，MACD金叉，反映市场极度恐慌。中概股延续空头排列，BABA、PDD、JD价格均低于所有均线，RSI 36-41，动能偏弱。商品外汇方面，黄金期货4321美元，跌破20/50/200日均线，RSI 33.4逼近超卖；WTI原油94.84美元，平盘整理；美元指数DXY 100.12，持续站稳所有均线上方并接近52周高，MACD金叉，偏多；10Y美债收益率4.54%维持多头。整体技术面除美元、美债、VIX偏强外，其余资产普遍承压。
VIX当前21.51，高于20日(17.12)、50日(18.99)和200日均线(18.42)，趋势bullish；RSI 65.7，MACD金叉（-0.3953 > -0.7728），近1日暴涨39.68%，近5日涨40.4%，显示市场恐慌情绪急剧升温，技术指标呈现明确偏上行结构。
DXY报100.12，高于20日(99.13)、50日(98.91)和200日均线(98.61)，趋势bullish；RSI 66.6，MACD金叉（0.2939 > 0.1874），信号显示“接近52周高”和“多头排列”，近5日涨0.93%，美元指数技术面延续强势上行。
BTC现价63,066，远低于20日(70,988)、50日(75,588)和200日均线(78,357)，空头排列；RSI 26.0处于超卖区，MACD -4037.45远低于信号线-2968.23，绿柱持续放大，近5日跌1.48%，技术面极度承压，偏下行态势明确。
TSLA报391，低于20日(425.87)和200日均线(414.14)，50日均线(395.29)仅略高于现价，触发空头排列；RSI 40.4，MACD死叉（4.143 < 8.607），近1日跌6.56%、近5日跌10.28%，技术信号普遍指向下行。
SPY现价737.55，低于20日均线(746.29)但高于50日(713.51)和200日均线(683.93)，趋势bullish；RSI 49.4中性，MACD 9.98略低于信号线12.06，接近死叉但未确认，近1日跌2.58%、近5日跌2.5%，多空力量平衡，技术状态中性。
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
过去走势不代表未来表现，本报告仅依据公开行情数据计算出的技术指标进行客观描述，不构成任何投资建议或预测。所有结论仅供技术指标解读参考，投资决策需结合其他信息独立判断。
Despite Trump’s calls for calm, IDF says it struck military targets in Iran in response to attacks by Tehran Israel strikes Iran despite Trump plea as Middle East crisis threatens to escalate The Israeli military said Monday that it had detected a new barrage of missiles launched from Iran, the sixt
中文摘要 尽管特朗普呼吁冷静，以色列国防军仍对伊朗军事目标发动打击，以回应德黑兰的袭击。中东危机加剧，以色列与伊朗再次交火。
Iran fired missiles at Israel in retaliation for an Israeli attack in Lebanon. Hours after the Iranian launches, the Israeli military said it had struck military targets in Iran.
中文摘要 伊朗向以色列发射导弹，报复以色列在黎巴嫩的行动。数小时后，以色列军方称打击了伊朗境内军事目标。这是自4月停火以来双方首次直接交火。
Follow live Get our breaking news email, free app or daily news podcast Is Australian music at risk of extinction? Here’s what the data tells us The music that charts in Australia has changed considerably over the past couple of decades – rock is out, country is in and old tracks are new again. Beca
中文摘要 澳大利亚新闻：安德鲁·哈斯蒂指责波琳·汉森「美国优先」。此外，数据显示澳大利亚音乐市场格局变化，摇滚衰落，乡村音乐兴起。
Trucks with billboards depicting phrase alongside Jacinta Allan wearing a black pointed hat have been seen around Melbourne for about six weeks Follow our Australia news live blog for latest updates Get our breaking news email, free app or daily news podcast Julia Gillard and Anthony Albanese have j
中文摘要 澳大利亚前总理吉拉德和现任总理阿尔巴尼斯谴责针对维多利亚州州长阿伦的「赶走女巫」广告牌运动，称其「性别歧视」。
Residents warned not to enter damaged homes or other buildings due to threat of aftershocks following 7.8 earthquake A magnitude-7.8 earthquake shook part of the southern Philippines early on Monday, collapsing buildings and killing at least 15 people. “Many buildings were affected, but I cannot enu
中文摘要 菲律宾南部棉兰老岛发生7.8级地震，至少15人遇难，建筑物倒塌。当局警告居民注意余震风险，勿进入受损建筑。
Markets in South Korea and Japan slid after a rally in tech stocks in recent weeks.
中文摘要 受中东冲突升级影响，亚洲科技股在近期连续上涨后暴跌，韩国和日本股市大幅下滑。
A volunteer diver has described shaking as he filmed his encounter with an endangered Great White shark between Tunisia and Sicily.
中文摘要 一名志愿潜水员在突尼斯与西西里岛之间的海域拍摄到濒危大白鲨的罕见画面，他描述自己拍摄时浑身颤抖。
A volunteer diver has described shaking as he filmed his encounter with an endangered Great White shark between Tunisia and Sicily.
中文摘要 一名志愿潜水员在突尼斯与西西里岛之间的海域拍摄到濒危大白鲨的罕见画面，他描述自己拍摄时浑身颤抖。
Police and hunters in Utsunomiya, 100km north of the capital, resume their search for animal that is not usually seen so close to Tokyo A city in Japan has closed all its 94 primary and secondary schools after a bear was spotted in the municipality for the first time. Officials in Utsunomiya, a city
This blog is now closed – our live coverage of the Middle East crisis continues here Donald Trump also aggressively pushed back against claims that he broke a key campaign promise to keep the US out of new foreign conflicts. “Well, well, first of all, I didn’t guarantee no war,” Trump said during th
中文摘要 据报道，特朗普告诉内塔尼亚胡不要报复伊朗。特朗普还反驳了「违背竞选承诺让美国卷入新冲突」的说法，称「我从未保证没有战争」。
Media outlets argue man’s name should be made public and police acting to protect him from ‘embarrassment’ Follow our Australia news live blog for latest updates Get our breaking news email, free app or daily news podcast A Queensland magistrate has suppressed the name of a man with a “high public p
中文摘要 昆士兰一名地方法官下令隐藏一名「高公共形象」男子的姓名，该男子涉勒索案。媒体认为应公开姓名，警方则称此举是为避免其「尴尬」。
中文摘要 最新消息：伊朗与以色列冲突持续，暂无详情。
Meet the rights defender charged with protecting the country’s fighters — not from Russia, but from abuses by their own commanders.
中文摘要 一名人权捍卫者负责保护乌克兰士兵免受本国指挥官虐待，而非来自俄罗斯的威胁。
Latest exchange of attacks threatens to drag Middle East into regional war and comes after US president said ‘I call all the shots’, not Benjamin Netanyahu Middle East crisis – live updates Israel launched airstrikes on central and western Iran on Monday in apparent defiance of Donald Trump after he
中文摘要 以色列无视特朗普呼吁，对伊朗中部和西部发动空袭。最新的攻击可能将中东拖入区域战争，此前特朗普称「我说了算」。
From Iranian naval ports to US military bases across the Gulf, Al Jazeera shows 15 sites before and after the attacks.
中文摘要 半岛电视台发布卫星图像，展示美以对伊朗战争造成的破坏，包括伊朗海军港口和海湾美军基地等15处地点的前后对比。
Take private deal comes as UK ingredients group struggles with subdued consumer spending
中文摘要 英国食品配料公司Tate & Lyle同意接受美国竞争对手Ingredion以27亿英镑（约36亿美元）的收购，私有化交易发生之际该公司正面临消费者支出疲软。
Markets in South Korea and Japan slid after a rally in tech stocks in recent weeks.
中文摘要 亚洲科技股暴跌，此前几周曾创纪录上涨，韩国和日本股市下滑，中东局势再度紧张。
Israel and Iran exchanged missile strikes on Monday despite President Donald Trump’s calls for both sides to halt the fighting and give peace talks a chance to succeed. (Source: Bloomberg)
中文摘要 以色列与伊朗6月8日互射导弹，无视美国总统特朗普呼吁双方停火并推动和平谈判。
“Bloomberg: The China Show” is your definitive source for news and analysis on the world's second-biggest economy. From politics and policy to tech and trends, Yvonne Man and Stephen Engle give global investors unique insight, delivering in-depth discussions with the newsmakers who matter. (Source:
中文摘要 彭博“中国秀”节目6月8日讨论股市暴跌与AI热潮赢家，为全球投资者提供中国政治、政策、科技等深度分析。
Ingredion Inc. agreed to acquire UK-based rival Tate & Lyle Plc for £2.7 billion ($3.6 billion), marking another famous brand leaving the London stock market.
中文摘要 美国Ingredion同意以27亿英镑（36亿美元）收购英国Tate & Lyle，这是又一知名品牌离开伦敦股市。
Steep declines in tech-related stocks follow Wall Street rout on Friday
中文摘要 全球股市下滑，韩国市场领跌，科技股大幅下跌，此前华尔街上周五遭遇抛售。
German factory orders fell more than expected in April, adding to concerns that Europe’s largest economy could contract in the second quarter due to the Iran war and the surge in energy costs.
中文摘要 德国4月工厂订单降幅超预期，加剧对欧元区最大经济体二季度可能萎缩的担忧，原因是伊朗战争和能源成本飙升。
The scheme aims to tackle the "growing challenge" of young people not in employment, education, or training.
中文摘要 英国零售公司M&S启动针对1000名年轻人的新培训计划，旨在应对青年失业、失学、失训的挑战。
A group of high-profile private credit firms are lending €1.1 billion ($1.3 billion) to French business software firm Cegid, in a show of confidence in a sector that’s been rattled by an AI-induced selloff.
中文摘要 Arcmont和Ares领衔一批知名私人信贷公司向法国商业软件公司Cegid提供11亿欧元（13亿美元）贷款，显示对受AI抛售影响的软件行业的信心。
Italy’s largest bank says takeover would result in significant cost savings
中文摘要 意大利最大银行联合信贷(Intesa Sanpaolo)以306亿欧元出价竞购西雅那银行(Monte dei Paschi)，此前一天竞争对手Banco BPM已提出合并要约。
Intesa Sanpaolo offered to buy Banca Monte dei Paschi di Siena for €30.6 billion ($35.3 billion). The Milan-based bank will offer 1.6 of its own shares and €1 in cash for each Monte Paschi share. The bid comes the day after Banco BPM pitched its own merger with Monte Paschi, considered to be the wor
中文摘要 联合信贷(Intesa Sanpaolo)出价306亿欧元（353亿美元）收购西雅那银行(Monte dei Paschi)，每股置换1.6股Intesa股票加1欧元现金，此前一天Banco BPM刚提出合并。
Insight with Haslinda Amin, a daily news program featuring in-depth, high-profile interviews and analysis to give viewers the complete picture on the stories that matter. The show features prominent leaders spanning the worlds of business, finance, politics and culture. (Source: Bloomberg)
中文摘要 哈斯琳达·阿明主持的“洞察”节目6月8日分析AI热潮出现裂痕，亚洲抛售加剧，战争风险上升。
29 回复 · 程序员 节点
9 回复 · 程序员 节点
9 回复 · Apple 节点
11 回复 · Apple 节点
19 回复 · 程序员 节点
14 回复 · 程序员 节点
17 回复 · 程序员 节点
36 回复 · Apple 节点
11 回复 · Apple 节点
42 回复 · Apple 节点
之前写了两次作文没过，估计字太少了…，现在新规来了就注册了，终于不是游客了 223 个帖子 - 212 位参与者 阅读完整话题
各位佬，因为我个人精力的原因无法保证 sla，也有佬友评论吐槽过这一点，所以暂时取消 vip 订阅，另外说明一下，vip 订阅只是用来提高 rpm的，并不是充值，无法保证 sla 自然要取消，但是已经订阅的不受影响哈，其实各位正常使用根本就不用订阅 vip，普通分组的 rpm 限制为 10 应该都已经够用啦。 https://shop.aini8.com 活动的原因并不是因为 bug team 哈，已经进展好几天了，原因是我们要进行架构升级，升级之前就一直有这个活动，各位佬看清楚小店的商品说明再购买～ 38 个帖子 - 29 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 AI PPT赛道终结者，史上最最最强PPT Skill！！！ 使用GPT生成豪华的图片格式PPT，然后转换为完全可编辑的PPTX文件，效果几乎完美。 本技能是个技能包，包含以下3个技能，可以拆分使用和优化： 技能 作用
哎，感觉天都塌了，现在36岁，职业上刚爆发中年危机，随时有失业风险。我家孩子1岁11个月。疑似重度自闭症。 典型症状：应名差，不找人。不和小朋友玩，蝴蝶手+刻板语言+情绪随时失控+转圈，听不懂指令，大运动、精细运动都不会。什么玩具都不会玩，只会咬，碰几下一扔，比我家小的孩子啥都会了，和同龄的一比感觉完全是降为打击。媳妇也是年初3月份被裁员。去年还欣欣向荣的家，转眼就支离破碎了。想带孩子看看，网上都推建北京大学第六医院 郭延庆，挂了1周没抢到号，某鱼上2800一个号，实在承受不起，佬们有没有什么好的挂号办法，求分享，感激不尽 编辑一下，感谢大家的安慰和帮助。我现在已经闲鱼2800待抢了，但是不保
每天都要来L站逛逛，今天终于升3级了 在这一个多月里真是学到了不少东西，十分敬佩各位佬们无私的分享自己的一些经验与技术，以及感谢佬友们对各种问题的答疑 33 个帖子 - 31 位参与者 阅读完整话题
细品，这高考作文绝了 52 个帖子 - 51 位参与者 阅读完整话题
楼主不是计算机相关背景的，大学本科学的是艺术管理（这是个艺术专业），但是毕业之后还是阴差阳错的来到了一家AI公司工作。楼主在大学还没毕业的时候非常喜欢AIRP，当时25年上旬吧，就非常想给自己手机qq上能随时整一个可以陪我聊天的bot，结果在b站刷到了一个astrbot的视频，命运的齿轮就此开始转动…… 在这之前我想讲讲我这两年的经历，非常的平淡，也没什么人听我说，但我真的太高兴了这个晚上，我必须要写出来点什么。当初我大三大四的时候还是文心一言的时代（2024），国内AI都特别笨呢，楼主那会不喜欢这个专业的对口工作，当时也不太关注AI，但是有一个AI相关的工作找上我来了，那我必须得去啊。是这样
这个分组坊间传得很神秘，其实也没传闻中的那么神奇，今天小编就给大家来揭秘一下。 简单来说，这是一个分组，进入这个分组后账号将获得以下能力： 独特的角标和变色头衔 多一个目前空白的版块 降低回帖最少字数限制 无限期编辑自己的帖子 每30天可改一次用户名 每3天生成1个邀请链接 可以看到，基本上除了第一个，都是需要额外消耗不少服务器资源的小能力。有那好提问的佬友会说：特么改个用户名也消耗服务器资源？这位看官，还真叫你给说着了，看看 discourse 官方怎么说的： 好了，以上就是对这个分组的揭秘介绍。如何开放这个分组，目前还没想好。不过，除了管理和版主之外，bohe 刚才也进了。 273 个帖子
新增了一些grok模型 grok-4.3-console,grok-4.20-multi-agent-xhigh,grok-4.3-high,grok-4.20-multi-agent-console,grok-4.20-fast,grok-build-console 还有minimax的m3 MiniMax-M3 ps：签到站应该快修好了0.o，稍安勿躁 52 个帖子 - 51 位参与者 阅读完整话题
已严肃处理 79 个帖子 - 79 位参与者 阅读完整话题