“车企牌”电池相继上车，不只是PK宁德时代

“车企牌”电池正在不断涌现，并交付上车。鸿蒙智行推出了巨鲸电池，理想汽车推出了理想牌电池，不久前的7月30日晚，小米正式对外发布了龙甲电池。 车企正试图通过自有电池品牌，将电池的每个技术点攥在手中，并抢占用户心智。 这些汽车品牌的统一做法，就是“穿透式管理”，从电芯到电池包的技术和性能标准，车企全部主导定义，甚至部分参与电芯材料开发，并对核心供应链和生产制造环节全流程介入和监管。换句话说，“车企牌”电池是车企技术和标准的全面贯彻。 “甚至电池厂的核心岗位招聘，车企都要参与面试把关。”有行业人士透露。 这与过往的动力电池合作有着明显区别。据36氪了解，长期以...

自研高性能分布式解决方案，「芯晓科技」将芯片电源签核周期从几周缩短至几天 | 水下项目

一枚先进制程芯片从设计、验证，到流片、制造，需要经过上百个环节。其中电源签核（Sign-off）是设计流程最后的、也最重要的一环，相当于建造一栋大楼时对全楼供电系统进行竣工验收。 具体来说，电源核签是通过对电源网络的上亿个节点进行整体分析，测试供电网络的功耗、电压降与电迁移等表现，从而决定芯片能否顺利进入量产流片。一旦结果出错，不仅导致芯片性能不达标，甚至会造成流片失败。 由于芯片电源网络是由大量金属线组成的互联结构，因此电源签核时必须将其作为一个整体进行分析，“相当于求解一个维度高达数亿的稀疏矩阵方程，完成一次需要大量时间。”「芯晓科技」总经理谢卓解释道。 而随...

从追番到下单，国风IP正在重塑硬核数码消费

当国风联名吹到硬核数码 这个夏天，《八仙！》成了暑期档最大的黑马。凭借反套路的剧情设计与极具东方美学的画面一路狂飙，斩获豆瓣8.2的评分，预测总票房达到20亿。 想不到的是，电影里那只随手捏出来的“红土鸭子”，竟也成了今年夏天受年轻人追捧的玄学单品。 在官方授权店里，这只鸭子上架没多久，就售罄下架。没抢到的网友们也被逼成了野生手艺人，有人用红土陶复刻同款，有人用3D打印自制平替，也有人花两小时用拼豆拼了个2D的鸭子出来。 社交平台上，催促官方赶紧售卖周边的声音也此起彼伏， “急切想花钱但是花不出去”，“八仙你会做周边吗急死我了”……  图源@...

Replit Ambient Intelligence 每帧生成设计建议

Ambient Intelligence 会在每一帧旁边显示建议卡片，每张卡片都代表你设计的一个不同方向。点击你喜欢的那张，即可看到它生成一个新帧。 你再也不会不知道下一步该做什么了。 立即在 http://replit.com/design 试用（AIHOT源：X：Replit (@Replit)）

Replit 环境智能：免提示词自动生成设计

你无需提示词，也无需设计语言。 环境智能（Ambient Intelligence）会在每个画面旁显示建议卡片，每张卡片都指向你的设计的一个不同方向。点击你喜欢的那张，即可看到它生成一个新的画面。 你再也不必纠结下一步该做什么。 立即在 http://replit.com/design 体验。（AIHOT源：X：Replit (@Replit)）

MiniMax-H3 通过 MLX 移植可在 Apple Silicon 上运行

MiniMax 发布 MiniMax-H3，一个可接受文本、图像、音频和视频并生成最长 15 秒带音频视频片段的通用全模态生成系统。Python 包 PipeNetwork/minimax-h3-mlx 将其移植到 MLX，支持 Apple Silicon 运行。作者在 M5 Max MacBook Pro 上实测，下载约 115 GB 模型文件，视频生成耗时不到 45 分钟。（AIHOT源：Simon Willison 博客）

Soup v0.72.4：在4 GB显存笔记本GPU上微调8B模型

Soup 推出 v0.72.4，支持在配备 4 GB 显存的笔记本 GPU 上通过 QLoRA 微调 8B 模型，无需 SSH 或云服务。（AIHOT源：Hacker News 热门（buzzing.cc 中文翻译））

OpenRouter 上线 FLUX 3 Video 统一多模态模型

@bfl_ai 的 FLUX 3 Video 现已在 OpenRouter 上向所有人开放。 一个统一的视频、音频、图像和动作预测多模态模型家族。严肃、有趣、创意、真实、电影感，随你所需。基于统一架构联合训练。（AIHOT源：X：OpenRouter (@OpenRouter)）

SpecForge v0.3.0 发布：统一解耦与共置投机解码栈，新增开放 SpecBundle 草稿模型

SpecForge v0.3.0 将目标模型推理与草稿模型训练分离，支持 EAGLE3、EAGLE3.1、P-EAGLE、DFlash、Domino、DSpark 等多种投机解码算法，并统一在线、离线与解耦工作流。（AIHOT源：LMSYS：Blog（Chatbot Arena 团队））

GitHub 如何用堆叠式 Pull Request 拆解 AI 生成的巨型代码

GitHub 介绍用堆叠式 Pull Request（stacked PR）解决 AI 编码智能体生成巨型代码难以审查的问题。通过将 1，000+ 行的大 diff 按数据、API、接线、UI 拆成 L1-L4 四个独立分层，每层可分配不同审查者。（AIHOT源：GitHub Blog）

Google Cloud API Gateway 推出统一模型路由功能，支持 Gemini、Claude 与 OpenAI OSS-GPT

Google Cloud API Gateway 新增模型路由功能（Public Preview），开发者可在 OpenAPI 3.x 规范中配置虚拟模型名到后端目标的映射，无需硬编码端点或管理开源代理。网关作为无服务器入口层，接受标准 OpenAI 兼容请求，自动将负载转码为目标模型的原生格式并动态路由流量。（AIHOT源：Google Developers Blog（RSS））

Google Cloud 推出 Database Operations Agents，实现自主数据库管理

Google Cloud 在 Agentic Data Cloud 发布中推出两款 AI 数据库智能体：Database Onboarding Agent 负责 Day 0 的配置与部署，Database Observability Agent 负责 Day 1/2 的监控、故障排查与维护。（AIHOT源：Google Cloud：Databases（RSS））

商汤 SenseNova U1 开源：统一推理与图像生成

商汤发布开源模型 SenseNova U1，可在统一流程中同时进行推理与图像生成。其信息图模式可将单条提示词转为结构化幻灯片，交错模式则逐步生成图文内容，如演示六步画龙教程。模型已上线 SenseNova Studio、HuggingFace 及 GitHub。（AIHOT源：X：商汤 SenseTime (@SenseTime_AI)）

蚂蚁百灵发布Ling-3.0-flash开源权重

今天，我们发布了 Ling-3.0-flash 的开源权重。🎉 官方 BF16 和 FP8 量化版本现已可用，您可以根据自己的硬件、性能要求和部署需求选择最合适的版本。（AIHOT源：X：蚂蚁百灵 (@AntLingAGI)）

NVIDIA Alpamayo 2 Super 开放商用，面向 Robotaxi 与自动驾驶的前沿开源模型

NVIDIA Alpamayo 2 Super 现已开放商用，基于 Cosmos 3 Super Reasoner 构建，采用强化学习后训练，支持轨迹预测、因果链推理、元动作、自动标注及视觉问答等多任务输出。（AIHOT源：NVIDIA Blog（RSS））

Stop Calling Everything Impostor Syndrome: The Myth of "Just Push Harder"

Damn Popovers

Top 7 Featured DEV Posts of the Week

Interviewing off leetcode you already memorized isn't cheating, it's the job

Understanding Over Origin: The Missing Friction

Anthropic Engineering: An update on recent Claude Code quality reports

Over the past month, we’ve been looking into reports that Claude’s responses have worsened for some users. We’ve traced these reports to three separate changes that affected Claude Code, the Claude Agent SDK, and Claude Cowork. The API was not impacted. All three issues have now been resolved as of April 20 (v2.1.116). In this post, we explain what we found, what we fixed, and what we’ll do differently to ensure similar issues are much less likely to happen again. We take reports about degradation very seriously. We never intentionally degrade our models, and we were able to immediately confirm that our API and inference layer were unaffected. After investigation, we identified three different issues: On March 4, we changed Claude Code's default reasoning effort from high to medium to reduce the very long latency—enough to make the UI appear frozen—some users were seeing in high mode. Th...

Anthropic Engineering: Scaling Managed Agents: Decoupling the brain from the hands

Get started with Claude Managed Agents by following our docs . A running topic on the Engineering Blog is how to build effective agents and design harnesses for long-running work . A common thread across this work is that harnesses encode assumptions about what Claude can’t do on its own. However, those assumptions need to be frequently questioned because they can go stale as models improve. As just one example, in prior work we found that Claude Sonnet 4.5 would wrap up tasks prematurely as it sensed its context limit approaching—a behavior sometimes called “context anxiety.” We addressed this by adding context resets to the harness. But when we used the same harness on Claude Opus 4.5, we found that the behavior was gone. The resets had become dead weight. We expect harnesses to continue evolving. So we built Managed Agents: a hosted service in the Claude Platform that runs long-horizo...

Claude Blog: New in Claude Managed Agents: self-hosted sandboxes and MCP tunnels

Starting today, Claude Managed Agents can operate in a sandbox you control and connect to your private Model Context Protocol (MCP) servers. Both the sandbox where an agent executes tools and the services it reaches run within the established boundaries of your enterprise, under your security and runtime controls. The sandbox runs on your own infrastructure, or with managed providers like Cloudflare , Daytona , Modal , or Vercel to handle the compute and isolation for you. On the Claude Platform, self-hosted sandboxes is available in public beta and MCP tunnels in research preview ( request access ). Self-hosted sandboxes: keep agent execution within your perimeter A self-hosted sandbox lets a Claude Managed Agent execute tools on infrastructure you control or with a managed sandbox provider. Code execution, sensitive files, packages, services, and data stay within your enterprise perime...

Thibault Sottiaux: Given some of the results I'm seeing recently, it's pretty clear Codex is a g...

Given some of the results I'm seeing recently, it's pretty clear Codex is a good harness. But it will seem primitive in 2-3 months and we're about to go through another major evolution in how we use AI at the frontier. The next generation of models need more than your laptop.｜@thsottiaux · 热度 9389 · 2026-08-04

Thibault Sottiaux: Crazy thing about OpenAI is you just open your laptop, codex a PR into existe...

Crazy thing about OpenAI is you just open your laptop, codex a PR into existence and ship an improvement to 1B users. Just like that. https://t.co/YwjkNCN0Oa｜@thsottiaux · 热度 4980 · 2026-08-03

Thibault Sottiaux: Some fine folks apparently misunderstood, but the GPT-5.6 Luna price reductio...

Some fine folks apparently misunderstood, but the GPT-5.6 Luna price reduction by 80% is not a temporary stunt, it's permanent. Efficiency gains don't go away. Fortunately. https://t.co/043ZF9T7gL｜@thsottiaux · 热度 3665 · 2026-08-04

Thariq: I think a lot of people don't realize- if you connect a Claude Connector (e.g

I think a lot of people don't realize- if you connect a Claude Connector (e.g. your gmail, calendar, slack, etc), Claude Code will also be able to use them, including in Artifacts. https://t.co/FZ1iSHqL5m｜@trq212 · 热度 1322 · 2026-08-03

Amanda Askell: I don't agree with this part

I don't agree with this part. I think the takeaway should be that models (like humans) can behave in aligned ways while still causing harm, e.g. because they're given false information about their situation. There isn't a line between aligned and harmless: they're different axes. https://t.co/btojNfpU5a https://t.co/0wCg27I2rC｜@AmandaAskell · 热度 1202 · 2026-08-03

Guillermo Rauch: Next.js 16.3: faster, better, stronger

Next.js 16.3: faster, better, stronger. My highlights: 0️⃣ Faster dev & builds Shoutout to Fable for improving memory use, and the team for landing the incremental 𝚗𝚎𝚡𝚝 𝚋𝚞𝚒𝚕𝚍 cache. 1️⃣ Instant navigations are 🔥 The people wanted SPAs, they shall get SPAs. By turning on instant navs (which will soon be default), your agent is basically forced to make things fast. 2️⃣ Agent-native DX You can still ship "slow" navs if you want, e.g. by blocking navs on server-side data fetching ala PHP olden days. But even then, your agent will be given *choices*, like setting up a <𝚂𝚞𝚜𝚙𝚎𝚗𝚜𝚎 𝚏𝚊𝚕𝚕𝚋𝚊𝚌𝚔> for instant loading shells. Agents also get built-in versioned docs. This is our best release for agents by far, which will also make your upgrading journey very smooth. I've had an easy time bumping my projects to 16.3 and enabling instant navs, having my agent migrate everything without issues. And so much more. 16.3 is more cost-efficient to serve, both in self-hosting and serverless environments. Faster to type-check… Thank you to the 90 contributors to this release!｜@rauchg · 热度 641 · 2026-08-03

Guillermo Rauch: The new AI Gateway logs UI is delightful https://t.co/cpMCaVeTbj https://t.co...

The new AI Gateway logs UI is delightful https://t.co/cpMCaVeTbj https://t.co/rwzkJdf0FW｜@rauchg · 热度 270 · 2026-08-03

Peter Yang: My 6 biggest takeaways from @karan4d, @NousResearch co-founder, on getting th...

My 6 biggest takeaways from @karan4d, @NousResearch co-founder, on getting the most out of Hermes Agent: 1. The “personal” in personal agent is not the model It’s the agent’s memory of your conversations and the skills you’ve built with it. Karan barely notices when he swaps models because his Hermes already has so much context about him. 2. Try different Hermes personalities Use /personality to switch response styles. Try “teacher” when learning something new, “concise” when you want crisp answers, and even “kawaii” or “pirate” for fun. 3. Use separate agents to do the work and evaluate it Karan calls “you’re absolutely right” reward hacking because the model is optimizing for your approval. Have one agent do the work and a fresh agent with no previous context evaluate it independently for errors, weak assumptions, and missing evidence. 4. A self-improving agent needs to clean up after itself Hermes Curator tracks skill usage, marks inactive skills as stale, and moves long-unused ones into a recoverable archive. This prevents agent-created skills from piling up forever. 5. Intelligence should be a public good Karan believes people should be able to run Hermes locally or use it with whichever model works best. He worries about a future where only a handful of companies dominate AI and open source gets squeezed out. 6. Go weird and make your childhood dreams come true Karan uses Hermes to mod Chao Garden, the virtual-pet mode inside Sonic Adventure 2. He asked it to bring a location from Sonic Adventure 1 into the sequel and add features that he’s always wanted since childhood. 📌 Watch the full episode here: https://t.co/3E7IC0TzNr｜@petergyang · 热度 185 · 2026-08-03

Guillermo Rauch: This is why P̵L̵G̵ ALG will always be king for startups

This is why P̵L̵G̵ ALG will always be king for startups. Have p̵e̵o̵p̵l̵e̵ agents adopt your product first, then have the meetings, if needed. The companies that start with the meetings are likely not your ideal customers. https://t.co/2e67lbJjgm｜@rauchg · 热度 182 · 2026-08-04

DietrichGebert/ponytail

Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.｜JavaScript · ⭐95818 · Search 增补 · 近快照 +969星 · 创建 2026-06-12

stablyai/orca

Orca is the ADE for working with a fleet of parallel agents. Run any coding agent with your own subscription. Available on desktop, mobile and VPS.｜TypeScript · ⭐37388 · Search 增补 · 近快照 +802星 · 创建 2026-03-17

Panniantong/Agent-Reach

Give your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.｜Python · ⭐66453 · Search 增补 · 近快照 +797星 · 创建 2026-02-24

Leonxlnx/taste-skill

Taste-Skill - gives your AI good taste. stops the AI from generating boring, generic slop｜JavaScript · ⭐71840 · Search 增补 · 近快照 +770星 · 创建 2026-02-19

Graphify-Labs/graphify

Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.｜Python · ⭐102468 · Search 增补 · 近快照 +655星 · 创建 2026-04-03

In Memory of My Wife, Elise Cawley, with Thanks for 36 Wonderful Years

Xbox goes down. You can't play games you own on disc

Show HN: Simple algorithm and color space to generate diverse skin tones

Hello HN! I found that picking out plausible but diverse skin tones for my digital art and game development projects was kind of difficult, and I got curious about if there was a way to define a color

DeepSeek V4 Flash on a Single AMD MI300X

There Will Come Soft Rains (1950) [pdf]

AMD 2026 财年第二财季归母净利润 22.97 亿美元，同比增长 163%

IT之家 8 月 5 日消息，AMD今日发布 2026 财年（2026 财年 2025 年 12 月 28 日～2026 年 12 月 26 日）第二财季（2026 年 3 月 29 日～2026 年 6 月 27 日）报告： 营业总收入 ：115.36 亿美元 （IT之家注：现汇率约合 780.51 亿元人民币） ，同比增长 50% 毛利润 ：62.03 亿美元 （现汇率约合 419.69 亿元人民币） ，同比增长 103% 毛利率 ：54% 归母净利润 ：22.97 亿美元 （现汇率约合 155.41 亿元人民币） ...

Lucid 2026 财年半年报归母亏损 23.94 亿美元，同比亏损扩大 62.79%

IT之家 8 月 5 日消息，Lucid今日发布 2026 财年（2026年 1月 1日～2026年 6月 30日）半年报（2026年 1月 1日～2026年 6月 30日）及第二财季（2026年 4月 1日～2026年 6月 30日）报告： 2026 财年半年报 营业总收入 ：6.88 亿美元 （IT之家注：现汇率约合 46.54 亿元人民币） ，同比下降 100% 归母净利润 ：-23.94 亿美元 （现汇率约合 -161.95 亿元人民币） ，同比亏损扩大 62.79% 经营现金流 ：-24.08 亿美元 （现汇率约合 -1...

倒计时 1 天｜跨境电商 AI Workforce 大会明日重磅启幕

点击查看原文>

RAG 不够用了？揭秘纯向量检索的 3 个盲区！

点击查看原文>

知识库越建越大，为什么团队还在重复踩坑？

点击查看原文>

The Lua community needs to learn to move on

Soppo - Go, with the features it's missing

Security is Hard, Y’all

Liquid Glass effect for GNU/Emacs

Enabling the next iteration of the borrow checker on nightly

The Download: US robot restrictions, and ICE&#8217;s DNA grab

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Trump’s AI protectionism has come for robotics   —James O’Donnell  Humanoid robots usually elicit more cringe than awe: They stumble, kick children, and despit...

Trump’s AI protectionism has come for robotics

This story originally appeared in The Algorithm, our weekly newsletter on AI. To get stories like this in your inbox first, sign up here. Humanoid robots usually elicit more cringe than awe: They stumble, kick children, and despite advances are still worse at using their hands than my toddler. It...

The Download: reward hacking explained, and suspected Iranian cyberattacks

This is today’s edition of The Download, our weekday newsletter that provides a daily dose of what’s going on in the world of technology. Here’s why AI agents lie and cheat to reach their goals When two OpenAI models hacked into Hugging Face last month, they weren’t trying to make money or commit...

Elon Musk repeatedly one-upped his execs on SpaceX&#8217;s first earnings call

Musk kept inflating the already-big promises being made by SpaceX CFO Bret Johnsen and Gwynne Shotwell on the company's first call.

Lucid&#8217;s turnaround plan hinges on $1.4B in cash savings, robotaxis

Lucid's new CEO Silvio Napoli listed four must-win priorities, including the successful launch of its midsize EV, finishing a factory in Saudi Arabia, cutting expenses, and robotaxis.

Now you can securely link multiple phones to one Signal account

You can link more devices with one phone number on Signal now, including an Android phone or iPhone. Signal already supported linking PCs and iPads, but not additional phones. When you link a device on Signal, you can check and respond to new messages across all of your linked devices. You can al...

AMD&#8217;s datacenter business is booming while gaming takes a backseat

Driven by demand for AI capacity, AMD's data center revenue more than doubled year-over-year in its latest earnings report, reaching $6.7 billion. That's up from $5.8 billion in Q1, and jumping 107 percent from the $3.2 billion it reported for the same period a year ago. At the same time, AMD's g...

敬畏热度、保持热情：如何在夏季坚持跑步

无论你是初跑者，还是跑步老手，相信这份「夏季跑步指南」都能帮助到你。 查看全文

家庭饮品 DIY 指南（三）：常用固体物料推荐

往期文章：（一）工欲善其事，必先利其器（二）常用液体物料推荐之前说的液体物料只是把最常用的几类跟大家分享一下，其实还有非常多好喝的好用的液体物料无法一一说明，后边配方中如果需要会顺带介绍。这篇聊一聊固 ... 查看全文 本文为会员文章，出自 《单篇文章》 ，订阅后可阅读全文。

社区速递 152 | 派友的固定回购清单、治愈系果壳风铃与超轻薄迷你风扇

除了首页时间流和侧栏的精选展位，少数派Matrix社区还有很多优秀内容因条件所限无法得到有效曝光，因此我们决定重启Matrix周报，并在此基础上添加更多社区内容、作者投稿新玩意呈现给大家。上周社区速递 ... 查看全文

别再只会 `find` 了：Map 在前端业务里的真实用法

前言 很多前端同学知道 Map，但真正写业务时却很少用。 原因也很简单：接口返回的数据通常是数组，表单提交的数据通常是对象或数组，最后给后端的也是 JSON。大多数时候，数组和普通对象已经够用了。

FFmpeg 9.0 发布，代号 Lei，音视频处理再升级

咱们平时用的播放器、剪辑软件、直播工具，甚至浏览器，底层大概率都跑着同一个东西：FFmpeg。 它是音视频领域的瑞士军刀，几乎无处不在。 2026 年 8 月 4 日，它发布了 9.0 大版本，代号

LangGraph 状态机实战

本文介绍了LangChain和LangGraph在构建大模型应用中的关键作用，逐步从简单API调用升级为可编排的有状态Agent系统。

特朗普与卡塔尔埃米尔通电话 官员对美伊达成临时协议表示乐观

周二，美伊达成一项聚焦霍尔木兹海峡的临时协议的可能性似乎正在上升，卡塔尔表示一份提案已起草，而美国和伊朗官员均对达成恢复这条关键水道通航的协议表示乐观。

外交争端加剧 美国撤销巴西大使签证

美国国务院一位高级官员周二透露，由于巴西迟迟未批准华盛顿选择的驻巴西利亚使节，特朗普政府撤销了该国大使的签证。 不过这位高级官员表示...

大宗商品综述：油价下挫 铜价和金价上涨

油价跌至逾三周低点，有关霍尔木兹海峡重开的预期升温。铜价升至5月中旬以来最高水平，交易员关注美国铜库存激增，以及美国有关铜关税的决定。

报道：美国告诉AI公司不会对开放权重模型进行安全测试

路透社援引两位未具名知情人士报道称，美国告诉领先的人工智能公司，不会对开放权重AI模型进行安全测试。

8月5日外盘头条：美伊谈判预期乐观引发油价大跌 SpaceX、AMD财报均超预期 OPEC上月石油产量增加

全球财经媒体昨夜今晨共同关注的头条新闻主要有： 1、卡塔尔和美国释放积极信号 聚焦霍尔木兹海峡的美伊协议渐现曙光 2、加拿大自美国进口额创新高 受数据中心建设推动 3...

开源版Claude Science来了！零依赖、MIT协议，内置30+项科研Skills

北京大学&元空AI Agent联合实验室

数学家24小时驳回OpenAI攻破的猜想！“AI证对了每句话，但已跟原猜想无关”

第二天，就有一篇人类论文回应：AI提出的反例不成立

开发者苦 “造轮子” 久矣，HarmonyOS 7 正在抹平系统能力的接入鸿沟

Skill和Agent也能被封装调用