개발 도구

sqlelf 用 SQL 声明式探索 ELF 文件，作者提出用 SQLite 替代 ELF 作为可执行格式的激进构想。

SDAD 论文将 Spec-Driven Agentic Development 形式化，主张规范质量是 agent 自主交付的执行燃料。

Terminal Agents 综述以终端执行为组织视角，建立七维终端能力画像，系统梳理命令行环境中的 LLM agent。

llm-anthropic 0.27 发布，兼容 anthropic v1.0.0 Python 库（httpx 迁移至 httpx2）。

GPT‑5.6 is now available in Kiro, helping developers plan, build, review, and test software with better price-performance.

Sounds quite interesting, a big IDE provider optimizing for local AI with their coding harness. Especially that they picked Qwen3.6 over Qwen3.8 because of the thinking needs. Haven't read the full article yet, but sounds really cool.

OCR It 是一个 Chrome 扩展，可对无法复制的分页文档批量截图并本地 OCR，输出纯文本供 LLM 使用。

arXiv:2608.20342v1 Announce Type: new Abstract: Large language model (LLM) coding agents start each session with an empty context window, discarding accumulated knowledge from prior work. We present PrimeAgentOrchestrator (PAO), a system that spawns new instances of Claude Code -- Anthropic's terminal-based coding agent -- pre-loaded with relevant memories compiled from the user's existing personal databases. At spawn time, PAO queries two independently-operated memory backends in parallel (a Po

Try it here: Model: Qwen 3.8 27b Q8_X_KL Unsloth Hardware: 3 x RTX3090 Harness: DeepSeek Harness Prompt: /goal I want you to create a **JavaScript + Node.js WebGL project** that renders a highly realistic real-time ocean in the browser. Use **JavaScript only, no TypeScript**. You may use WebGL2, GLSL, and Three.js. The ocean should include realistic waves, vertex displacement, Fresnel reflections, sun highlights, sky/environment reflection, foam/whitecaps, horizon treatment, atmospheric effects,

《Coding expertise is going to collapse from AI reliance》引发热议，多数评论认同 AI 依赖正在削弱编程技能，但也有人视其为技术演进必然。

评论普遍认同AI依赖正削弱编程技能，但也有人认为这是技术演进的必然，类似计算器或编译器的影响。

Article URL: Comments URL: Points: 80 # Comments: 39

At TechCrunch Disrupt 2026, Replit CEO Amjad Masad will share his perspective on the future of programming and Replit's role in developing it.

arXiv:2608.20389v1 Announce Type: new Abstract: A production agent harness must discover and rank, from a growing library of skills, the one most appropriate for a user's task. At small scale this selection happens in context: the LLM planner chooses among skill representations exposed in its system prompt, without an explicit embedding-based retrieval step. We treat this in-context selection as the small-N counterpart to embedding-based skill retrieval at scale, and present a case study of how

In this tutorial, we explore a LabPlot-inspired scientific data analysis workflow in Python while preserving the structure and terminology of LabPlot’s aspect tree, analysis kernels, plotting system, and project model. We build reusable components to import tabular data, compute descriptive statistics, smooth and differentiate signals, perform Fourier analysis and filtering, detect peaks, integrate curves, reduce […] The post Scientific Data Analysis with LabPlot in Python: Signal Processing, Sp

I commented on another Qwen 3.8 27B post that I was frustrated getting anything to work. You all gave some great comments. I nuked openwebui and straightened out my llama.cpp docker config. 1 hour of work and I have a model I can chat with, connected to my HomeAssistant server, which I have already updated dashboards with a short prompt and a screenshot (wtf vision built in?) Guess all I needed was the right push. I bought several GPUs in 2023 in impulse purchases for Folding@Home, but have alwa

Article URL: Comments URL: Points: 144 # Comments: 27

The memory layer that decides what's worth remembering Discussion | Link

The AI runtime that remembers, learns, and acts everywhere Discussion | Link

Our devs got their hands on it a few days ago. One wired it into Codex to compare with GPT Luna, our usual workhorse right now for its cost effectiveness. Another tried it out on one of our OCR pipelines. It's comparable to Luna for coding and ***OCR quality appears to be better than Gemini 3.5 Flash Lite***. That's huge. We pay a ton of money for OCR. This is the first local model that feels like more than a toy. It's truly as capable as the frontier models from a year ago. For the first time e

评论区普遍认可Qwen 3.8 27B在逆向工程等任务中的实用性和持久性，但也有人认为其他模型表现更优。

开发者微调 Gemma 4 12B 以适配 16GB 显存，工具调用能力提升 2.7 倍，针对 agentic coding 和 CLI 场景优化。

Most YouTube benchmarks only show trivial tasks like generating landing pages or simple Three.js games. Is a local model like Qwen 3.8 27B actually capable of real-world systems programming—such as building GTK4 or Qt 6 applications in Rust or C++ with external libraries? Specifically, if I look up the exact terminology in the online docs and then prompt the AI to inspect the cloned repo, can it implement the feature cleanly?

MartyPC 是用 Rust 写的跨平台早期 PC 模拟器，以高精度著称，Web 版已支持 IBM PC 5150 等机型。

评论区主要认可MartyPC的准确性和Rust语言优势，但也有人认为标题强调Rust多余，且存在键盘和硬件支持不足。

pgrust 作者分享如何在 5 微秒内完成 JIT 编译，借助 AI 直接面向汇编生成代码，为数据库 JIT 提供新思路。

评论区普遍认可JIT编译的趣味性与技术价值，但也有人认为其安全性不足、非真正优化。

In this tutorial, we explore how to design production-grade safety for LLM-based applications using the NeMo Guardrails framework. We move beyond simple prompt filtering to implement a layered architecture, featuring deterministic PII redaction, retrieval filtering, output masking, and policy-based tool gating. By integrating stateful multi-turn evaluation and detailed activation tracing, we demonstrate how to build an auditable, secure, and cost-effective AI assistant capable of managing sensit

I was just curious how the new qwen3.8:27b does on a hard C to HTML porting job against Opus 5 in a default Claude Code. The job: my fun side project is a procedural shooter in a single C file. Port it to a single-file html / three.js with one bot. One prompt, no follow-ups, no help from me. game.c is 2.1 MB, roughly 600k tokens of C, so it doesn't fit in the window and the agent has to walk the file and work out what matters. Setup: qwen3.8:27b in FP8 on vLLM, FP8 KV cache, full 262144 context,

用户从 Windows 切换到 Linux 后，llamacpp 换 vllm 推理速度提升 30-50%，评论区归因于驱动和内存管理差异。

Prior to Fable, it felt silly to waste too much time improving your coding harness or context strategies. A new model would arrive at the same price (or cheaper!) and paper over most of your problems. But then Fable landed. It was (and still is!) incredible . But the cost was so high and Opus was good enough (as was 5.6, K3, and even GLM) for most of the code we needed. So we started to think about what work went where. — Drew Breunig , Fable & The End of the Free Lunch Tags: drew-breunig , anth

Vercel 与 Ora 推出免费工具 Is Agentic，用 118 项检查审计网站对 AI agent 的可发现性与可用性，提供 CLI 和 MCP server。

deepDoctection 教程展示端到端文档智能管道：布局分析、DocTR OCR、表格提取与结构化 JSONL 导出，面向 RAG 工作流。

I don't know about you guys, but Deep-seek harness is insane. It's not focused on being a coder agent, it's webUI made it very easy to just checkin from time to time, and the best part? Why it's better than Hermes? It wasn't frustrating at all to setup. ZERO. NADA. Progressive setup is such an improved UX. Why? Because I got deepseek to integrate with SimpleX by simply asking it to. BY SIMPLY ASKING IT TO. NO WAITING ON A PR TO MERGE. No one telling me to RTFM, no need to google or search for co

Wrote a bit about the `unlikely` and `likely` macros in the Linux codebase. Great little exploration into tiny optimizations that everyone should know more about 😄

自托管社区讨论是否仍有人坚持纯 Linux + Docker Compose 方案，多数人认可其简单可控，但 UI 工具仍有存在价值。

I see so many people in here using Proxmox. I run Ubuntu + Docker, and my whole setup is just a list of folders with compose.yamls and supporting files (configs etc). It feels very manageable this way, fully declarative, easy to version control. Am I missing something, that makes people recommend and use Proxmox instead?

The key skill required to make productive use of coding agents is being able to confidently instruct them on how to make changes and then confidently verify that those changes have been applied in the correct way. Sometimes this involves reviewing every line of code they have written, but there are other ways to achieve that goal. Eyeballing every line of code has never been the most effective way to validate a chance to a piece of software. Tags: code-review , coding-agents , generative-ai , ag

And this was a debug session from hell, enormously helped by an AI doing much of the grunt-work. I'd like to call it my tireless helper, but the AI several times stated flat out that this was impossible and unsolvable and that we should just write a report about it. I suspect those things have been trained by people who may not be quite as stubborn as I am. But while the AI was ready to give up several times, it did keep adding debug code and analyzing it faithfully when I pushed. So credit wher

llm 0.33 发布：升级 OpenAI Python 库 3.x，embed 命令支持 --key，prompt -t 可重复组合模板。

评论区普遍质疑Anthropic暗中降低模型努力水平并存在计费不透明，但也有人认为这是官方测试且性能未受影响。

That is the second comparison and the last one. I will not be spamming again ;) Continuation from: That is one of my many tests I make comparing output quality. What is more interesting using a PI Agent results are much better than an Opencode using a Qwen 3.8 27b ?! Seems PI Agent is much better in the agent environment somehow... Not counting uses less tokens , do not have a hard limit of 32k output tokens, is faster, do not freezing, compressing context far less than Opencode. For instance if

Dan Luu 撰文称软件没有理由再慢下去，LLM 已将性能优化的成本降低数个数量级。

评论普遍认为软件变慢源于优先级、激励和工程权衡，而非技术能力；但也有人认为AI优化有限，慢软件仍将存在。

Munder Difflin 是一个在本地运行多个“克隆 agent”的 harness，评论区认为有趣但混乱难懂。

评论区普遍认为该项目有趣且具讽刺意味，但也有人认为它混乱无用或难以理解。

Level1Techs 论坛文章分析为何本地 LLM 感觉比基准测试笨，指出推理实现差异是关键。

I think agent-first chat interfaces will be a primary software modality and busy dashboard/UI will go away. I’m not sure who exactly wins it, but I want my knowledge to grow/go with me. A lot of the “knowledge” ie research, analysis, reasoning will be done by agents as the primary user. Our current notes tools & tasks management systems were built for humans… I don’t care what the 17th thing on my bug backlog is. I want to conduct agents that can execute for me and do great work. What I built Oz

An AI work agent across desktop, browser, and chat Discussion | Link

《A Friendly Introduction to Racket》以教程形式介绍 Lisp 家族语言，强调代码即数据与宏系统。

While studying Geometric Algebra I have built some interactive visualization to demonstrate how geometric transformations (rotation, scaling, translation) can be constructed by just composing reflections. Accepting reflection as the most elementary geometric operation was an eye opening moment for me. I think some of you might enjoy the interactive visuals. Comments URL: Points: 51 # Comments: 9

Libredesk is a self-hosted customer support desk for email and live chat. It's fully open source under AGPL, with no paid tier or separate enterprise build, No feature paywalls. Features: Email inbox and live chat widget, with both landing in the same agent inbox. Help center with collections, articles, search, and per-language content. Autonomous AI agent that answers from your knowledge base and hands off to a human when it can't answer. Agent copilot for drafting replies, summarizing conversa

Certain logos started standing out to me on LinkedIn as brighter/whiter than everything else around them. I dug in and found out this is accomplished by adding a gain-map to an existing JPEG, visible only on HDR screens like a newer MacBook Pro. LinkedIn is the only social network I've found that isn't stripping them out, but of course you serve them up on your own site. I worked with Claude Code to turn it into a little browser-based utility (no registration) and hope you find it useful! Commen

I used LM Studio Bionic with Qwen 3.8 27B Q3_K_S with 57k context. It took a staggering 63 hours to finish coding. After the first prompt "Create a beautiful, relaxing flight simulator in a single HTML page" taking 47.8 hours, it created an html file that showed the title screen that said "press any key" but pressing any keys won't advance the game. So I wrote on the second prompt "It saids press any key to begin. I press any key but it doesn't work." It ran for 15 hours. Now I can fly. No plane

llm 0.32.1 修复因 OpenAI Python 库弃用 httpx 导致的新安装故障，0.33 将切换到 httpx2。

After I released version 1.0, I figured I would have to do the rotations myself. So I sat down with ChatGPT and I didn’t get it to write the code, but I got it to educate me. With a patient, interactive tutor, I was able to finally do what I hadn’t by reading books and asking mathematician friends – I learnt how to use quaternions just enough to make the app work. So learning doesn’t stop just because I outsource a bunch of thinking to AI. It pushes me to learn more. I like that as an outcome. —

Bun 1.4 发布首个 Rust 重写后的稳定版，新增 Bun.WebView、Bun.Image 等 API，Node.js 兼容性大幅提升。

I tried this model yesterday, and it felt to me like the best one I've tried for a local model for interactive use; the responses and reasoning are very fast, and it actually performs agentic tasks well. The speed is phenomenal. I am running this on Ninfer for Windows

A quick feedback after a really major test: nearly 20 hours of non-stop goal-oriented work with Qwen3.8-27B Q6, running across an RTX 3090 and an RTX 3060. It maintained a speed of around 60–63 tokens/s throughout the session.

OpenRouter 上线匿名 stealth 模型 Ox Alpha，评论区普遍猜测为中国 GLM 系列模型，认可能力但质疑匿名做法。

评论区普遍猜测Ox Alpha为GLM系列中国模型，认可其能力但质疑匿名做法，也有人认为可能是西方模型。

This tutorial explores AutoFigure, a practical toolkit for generating professional scientific figures directly from text descriptions and research papers. We walk through setting up the environment, configuring an API-backed generation workflow, and converting complex document intelligence pipelines into publication-style diagrams, complete with custom reference styling and gallery exports. The post Building Agentic Document Intelligence Pipelines: Creating Scientific Figures with AutoFigure app

llm-openrouter 0.7 兼容 LLM 0.32，改用 OpenRouter 的 Responses API，并新增 Shell、WebFetch、WebSearch 三个服务端工具。

Claudette 是一个 Claude Code skill，用 Gemini CLI 将 Claude 的 BuzzFeed 式回复翻译成正常英文。

Antigravity agents now live inside your existing editor Discussion | Link

Run Claude Code, Codex or Opencode in cloud with your repo Discussion | Link

大厂工程师建议将自托管经历写入简历，认为其涵盖部署、监控、安全等真实工程问题的缩影。

Stop Making TUIs Thomas Ptacek advocates for building real native user interfaces for even the smallest of personal tools, because coding agents have reduced the cost of getting a usable-enough GUI up and running to almost nothing. I wrote about my vibe-coded bandwidth and GPU monitoring macOS task bar apps back in March , and I'm still using both of those on a daily basis. I'm not habitually knocking out real UIs for my other projects yet, but I'm running out of excuses! Thomas: If you haven’t

Kagi 新增自动移除付费墙链接的设置，评论区普遍赞赏其优于谷歌，但也有人希望保留优质付费内容或提供白名单。

评论区普遍赞赏Kagi移除付费墙链接功能，认为实用且优于谷歌；但也有人认为应保留优质付费内容或需白名单选项。

My Pro subscription expired today, they killed my access at 1pm local time. I'm now using Qwen3.8-27b w/ 5090m 24gb vram and pi to do everything i was doing in claudecode. The only downside is claudecode let me code without using my gpu, meaning I have to plan things now. Last night I had ChatGPT write up a prompt for a fancy aurora predictor for Canadians. I fed it to local pi and claude sonnet 5. They took about the same time, pi's app looked better, but claude's had better science. I asked th

I have been using omarchy on my tower since nearly a year now, shortly after it was released first. I really love the experience I am having with it but I still use my macbook for daly work, so I wanted to recreate a similar experience on it. Thats why I created omacosy, a setup for tiling windows, custom menu bar, some themes from omarchy, focus follows mouse, focus rings around windwos, some mac flavors with trackpad events and a custom mission control overview for your workspaces. I used Aero

Your personal AGI on your PC. Built to finish real work. Discussion | Link

Matt Pocock tells us about his /wayfinder skill, for greenfield projects or for when the way forward is unclear.

Ramp has launched its own AI model routing service, dubbed Router, that lets users and companies use and switch between various large language models via an API.

Huzzah 编辑器尝试用非自然语言方式与编码 agent 交互，解决长句描述变更的疲劳问题。

Slack is the new IDE

LEGO-RL 通过进程内 LLM 代理、沙箱编排与集成监控，将原生编码 agent harness 桥接到可扩展的策略梯度训练。

🤖LEGO-RL connects native coding-agent harnesses to scalable policy-gradient training via in-process LLM proxying, sandbox orchestration, and integrated monitoring, improving sparse MoE model performance across multiple harnesses.

Zoneless 开源 Stripe Connect 替代方案，声称将每月 9000 美元支付费用降至约 6 美元。

Programmable logic controllers (PLCs) run industrial plants, and large language models can already generate independent program organization units (POUs) for them. Whether such logic integrates into an existing PLC project and then runs correctly has been checked only in limited tests. We present SemaPLC, a project-grounded and verification-gated agent harness assembled from conventional tools but governed by a strict completion rule. Rather than stopping when the model judges its own output ade

🤖SemaPLC is a verification-gated agent harness that validates generated PLC logic through external compilation and live runtime execution, achieving higher verified pass rates than baseline methods.

评论区普遍认可Bun 1.4的Rust重写是AI辅助开发的重大成功，但也有人认为其功能膨胀、兼容性及AI依赖引发争议。

多数评论认同Claude 5输出冗长需清理，但也有人认为可换模型或调整提示词解决。

Last week I recorded an episode of the Talking Postgres podcast with Claire Giordano on the subject of "How AI is changing software development". We had a really great conversation. Here are a couple of my highlights from a lightly edited transcript (prompt to Claude: "very minor edits to remove disfluencies"). This is the latest version of an argument I've been trying to build about why sometimes it does make sense to talk about lines of code as an indicator of productivity with coding agents,

Turn spoken feedback into tasks your coding agent can run Discussion | Link

The agentic IDE for designers and programmers Discussion | Link

The new traffic peak caused "capacity failures" that led to the developer platform's recent outage.

Let non-engineers build secure internal tools Discussion | Link

「Don't paste the AI」引发热议：多数人反感直接粘贴 AI 回复，认为应加入个人判断；也有人认为 AI 回复更高效。

评论区普遍反感直接粘贴AI回复，认为这像LMGTFY般失礼；但也有人认为AI回复有时更高效，或能弥补个人写作不足。

Qwen3.8-27B 在双 3090 + vLLM + DFlash2 上单请求达 218 tok/s，prefill 1342 tok/s @ 10k，峰值显存 22.3 GB/卡。

llama.cpp pr #27342 adds dflash2, so i rented an rtx 6000 and ran the same four prompts through four decoding setups on qwen3.8 27B median results over the four tasks: baseline 47.4 tok/s mtp 114.7 tok/s dflash 99.3 tok/s dflash2 140.6. tok/s so on average 3x for dflash2 though i have to point out that it's far from a 3x gain some of the time, on one of the test it struggled to achieve a 1.5x gain, it really just depends on the task you give to the model the races are sped up in some places, so

Replit introduces Free Mode, powered by GPT-5.6 Luna, so anyone can turn ideas into working software without worrying about token costs.

KernelArc 多智能体框架在 H100/B200 上自主优化 GPU kernel，覆盖 BF16 GEMM、MoE backward 融合等负载。

arXiv:2608.17007v1 Announce Type: new Abstract: Agent Skills can specify procedural and resource obligations for tool use, and language models instantiate them as concrete programs. However, when models turn this guidance into code for existing tool interfaces, even a semantically correct program may load an entire input and exceed the memory available to one tool call. We present SkillEffect, a checked-lowering runtime for computations with a recoverable source relation, an audited bounded impl

The Git forge built for the age of coding agents Discussion | Link

Unsloth 发布 Qwen3.8-27B Dynamic v3 GGUF，同尺寸精度提升超 10%，1-bit 量化保留 77% 精度，8GB RAM 可跑。

Bun 1.4 Rust 重写引发社区担忧：作者批评其从工程亮点沦为 AI 驱动的持续空头承诺。

I have been using local models on/off for like 2 years or so but never really used them extensively because the closed ones were always much better. Once Qwen 3.8 27B was released I decided to give it another serious try. I configured Cline and ZooCode as VSCode addons, installed a few MCP servers and added one skill. When I used these tools with Deepseek V4 Flash - they do the job quite well (mostly Home Assistant configuration editing etc.) but it is still way worse than Claude Code/GitHub cop

作者用几何学与 CUDA 编程对随机岛屿照片进行地理定位，全程未使用 LLM 生成。

Framework says it's replacing some out-of-warranty AMD mainboards.

AI workspace for reproducible scientific research Discussion | Link

ByteDance Seed and Tsinghua AIR have released CUDA Agent, an agentic reinforcement learning system that trains a large language model to write GPU kernels that beat a compiler. The gap it targets is narrow but stubborn: frontier models already produce correct CUDA, they just produce slow CUDA. On KernelBench, the base model Seed1.6 passes 74.0% […] The post ByteDance Seed and Tsinghua AIR Introduces CUDA Agent: A Large-Scale Agentic RL System for CUDA Kernel Generation appeared first on MarkTech

Working on Mac, Linux, and Windows now. I include a simple GUI to find new models and get things built and set up. It is working quite well across a few models for me. The GitHub README and DESIGN.md files go into detail of the how/why and it's working remarkably well so far. Comments URL: Points: 38 # Comments: 6

NVIDIA 发布 TensorRT Model Connect 公开预览版，两条命令从 Hugging Face checkpoint 到原生 C++ 推理，无 ONNX 中间步骤。

Mojo🔥 is now open source Mojo🔥 is now open source The Mojo programming language has been promising an open source release since May 2023 . Last week they shipped their 1.0 and today they have followed through on that original promise, releasing the compiler and toolchain under an Apache 2 license. When Mojo first launched the stated goal was to produce a superset of Python, so existing Python code could be used to bootstrap their own ecosystem. That plan changed around August 2025 : Mojo may or

Asana used OpenAI Codex to replace an outdated testing system in two weeks, completing work expected to take five years for about $12K.

MyContext补上Agent的数据加工层

Two days ago I released a hyper-optimized Qwen3.8-27B inference engine for an RTX 3090 (82 tps single request, 672 peak) - yesterday's update took that to 99 tps single-user / ~1,000 tps at 64 concurrent. Since then I've focused on the single-request number, again without quality degradation. It's now ~114 tps at default sampling and ~124 tps greedy (real chat prompts, not random tokens), up from 90 / 98. What we had: - fp8 KV cache, lm_head + embed_tokens int8, fp16 recurrent state, int8 activa

You write a standard solution, just like on LeetCode, and run it through the CLI. It identifies the problem by ID or title, executes your code against local test cases, and shows the result. It currently supports around 1.4k problems and multiple languages, including Python, C++, Rust, Java, Go, TypeScript, Swift, and others. The project is still an MVP. System design, SQL, and concurrency problems are not supported yet, but support for more problem types is planned. Made in Haskell! Comments UR

Acadia 重新思考数据库编程，引入精确类型与可验证迁移，评论区认可设计但担忧许可限制。

评论区普遍认可其设计理念与功能亮点，但也有人认为它不过是SQL替代品或ORM，且许可限制令人担忧。

Cursor, known for its AI Code Editor, is launching a new code-hosting platform to rival developers' long preferred favorite, GitHub.

Turbovec 基于 Google TurboQuant 的 Rust 向量索引，1000 万文档语料从 31 GB 压缩到 4 GB，搜索速度超越 FAISS。

评论区普遍认可Turbovec性能与压缩效果，期待集成与WASM应用；但也有人认为文档不足、技术非新创。

Python Polars 速查表发布，评论区普遍认可其比 Pandas 更顺手，但也有人认为 DuckDB 更优且 pl.col 语法繁琐。

评论区普遍认可Polars比Pandas更顺手，但也有人认为DuckDB更优，且对pl.col语法繁琐有异议。

changed a db password in my .env, ran docker restart on the stack, then spent an hour convinced the db was corrupted because auth kept failing. turns out restart just brings the container back with the exact config it was created with. env is only read at creation. docker compose up -d --force-recreate fixed it in ten seconds. two years running this stack and never got bitten by it until now. what's the dumbest thing that ate an evening for you?

Passwords are less secure than passkeys, even if you use a password manager. Here's why and how to get started with passkeys.

Create and manage ElevenLabs voice agents in your chat Discussion | Link

Linux 7.3 合并 VRAM 超限性能优化补丁，评论区赞赏内核改进，但也有人关注 OOM 崩溃与 NVIDIA 支持问题。

评论区普遍赞赏Linux内核优化VRAM性能，期待改进落地；但也有人认为需关注OOM崩溃和NVIDIA支持问题。

Apple is once again overhauling App Store rules in the European Union, which the company says will resolve its "disagreements with the Commission over business terms and alternative distribution." As part of the changes, every developer that distributes apps will be moved to a single set of business terms, and digital transactions for apps distributed […]

Apple is simplifying its EU App Store fees, replacing its per-install fee with a 5% commission for apps distributed outside the App Store and making it easier for developers to operate alternative app marketplaces.

Build deterministic agents that actually follow your process Discussion | Link

Agent Led Growth: Get written into every customer's codebase Discussion | Link

A persistent terminal for Codex and Claude side by side Discussion | Link