[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-codex-bundles-libreoffice-ai-desktop":3,"topics-all":38,"news-related-407d6137-c0c6-4fde-84c1-4432b53e4cc4":57},{"id":4,"title":5,"summary":6,"content":7,"original_url":8,"source_id":9,"tags":10,"translations":24,"news_slug":31,"published_at":32,"created_at":33,"modified_at":34,"is_published":35,"publish_type":36,"image_url":14,"view_count":37},"407d6137-c0c6-4fde-84c1-4432b53e4cc4","Codex 把 LibreOffice 塞进桌面:1.7GB 工具栈暴露 AI 客户端的真实成本","Simon Willison 发现 OpenAI Codex 桌面端把 1.7GB 内容塞进 ~\u002F.cache\u002Fcodex-primary-runtime:完整 LibreOffice、Python、Node.js、Poppler、git 都在里面。Anthropic Claude 桌面端被曝捆绑 10GB 虚拟机。","Simon Willison(simonwillison.net)这个月初翻 ~\u002F.cache 时发现,OpenAI 的 Codex 桌面应用(已改名 ChatGPT)在缓存里塞了一个叫 codex-primary-runtime 的文件夹,1.7GB。展开:LibreOffice、Python、Node.js 一整套,再加 Poppler、git。`documents` 子目录的 skills 教 Codex 怎么调用这些二进制。\n\n## 为什么桌面 AI 助手需要 LibreOffice\n\nPDF 提取、docx 解析、表格读取,模型自己不会做,得有真工具来执行。Codex 把 LibreOffice 装进来,意味着它能在本地把 PDF 转 docx、把 docx 转 PDF,或跑 headless 模式做批量转换。Python 和 Node.js 是 skill 调用脚本的运行时底座。这是一台微型 AI 工作站,不是聊天客户端。\n\n## 1.7GB 和 10GB,两条不同的路\n\n但这事真正的对照,是 Anthropic 的 Claude Desktop。GitHub issue #22543(github.com\u002Fanthropics\u002Fclaude-code\u002Fissues\u002F22543)早就把这事挂出来:Cowork 每次运行都在 `~\u002FLibrary\u002FApplication Support\u002FClaude\u002Fvm_bundles\u002Fclaudevm.bundle\u002Frootfs.img` 写一个 10GB 的 rootfs 镜像,从不清理,删掉第二天又涨回来。8GB 内存的 Mac 跑 Claude Desktop,空闲 CPU 24%,几分钟涨到 55%,swap 持续爬升。临时绕过:`rm -rf` 三个缓存目录,任务速度提升 75%——但只是临时,内存泄漏本身没修。\n\nOpenAI 的 1.7GB 和 Anthropic 的 10GB,差别在打包哲学。Codex 走\"装齐原生二进制\"路线,体积小、启动快;Claude 走\"塞完整 Linux VM\"路线,功能隔离稳,但磁盘和内存开销高一个数量级。两条路指向同一结论:Agent 要在本地真正干活,光靠模型权重不够,它需要一整套传统工具栈。\n\n## Agent 范式正在反向 1990 化\n\n这背后是 Agent 产品的范式拐点。早一代 Copilot 把代码生成放在云端 IDE,本地只是轻量插件。新一代 Agent 要在本地读写文件、跑脚本、操作 Office 文档,直接调用本地工具比把所有能力都堆在云端模型里更现实。本地工具栈绕不开——用户交给 Agent 的是真实文档、真实 PDF、真实代码仓库,Agent 就得在本地把它们当二进制对象处理。\n\n代价由磁盘和内存买单——单应用 1.7GB,vm_bundles 10GB。开发者同开 Codex、Claude Desktop,30GB 是常态。再叠 Hugging Face、uv、npm,512GB 笔记本很快吃满。Willison 截图里,huggingface 一个就 356GB,uv 82GB。\n\n行业下一步要么学 Docker Desktop 做镜像去重和按需加载,要么学云端 IDE 把工具挪到远端容器。从 1.7GB 到 10GB,看见的是产品决策,看不见的是 Agent 范式正在把桌面软件吃回 1990 年代那种\"装齐全套工具\"的重客户端时代。\n\n所以下次装 AI 客户端之前,先看看它想在你硬盘里塞什么。","https:\u002F\u002Fsimonwillison.net\u002F2026\u002FSep\u002F1\u002Fcodex-libreoffice\u002F","4ca6d57c-7fa6-49f2-b7a1-f9c51b41bf2d",[11,15,18,21],{"id":12,"name":13,"slug":13,"description":14,"color":14},"40269b40-7942-4650-9672-ed2e6524d37a","ai-technology",null,{"id":16,"name":17,"slug":17,"description":14,"color":14},"23544f6a-eea1-4f05-aa8d-749ca862d5d2","anthropic",{"id":19,"name":20,"slug":20,"description":14,"color":14},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":22,"name":23,"slug":23,"description":14,"color":14},"42e59a88-7795-47dc-a334-ef1e72c24347","openai",[25],{"id":26,"lang":27,"title":28,"summary":29,"content":30},"c3f86e67-c96d-4fb8-95ee-505da89c01e6","en","Codex bundles LibreOffice into the desktop: 1.7GB toolchain exposes the real cost of AI clients","Simon Willison discovered OpenAI Codex desktop app stuffed 1.7GB into ~\u002F.cache\u002Fcodex-primary-runtime: full LibreOffice, Python, Node.js, Poppler, and git. Anthropic Claude Desktop has been exposed for bundling a permanent 10GB virtual machine.","Simon Willison discovered earlier this month, while poking around in ~\u002F.cache with OmniDiskSweeper, that OpenAI's Codex desktop app (since rebranded to ChatGPT) had stuffed a 1.7GB folder called codex-primary-runtime into its cache directory. Inside: a full LibreOffice, a full Python, a full Node.js, plus Poppler and git. The `documents` subdirectory contains skills that teach Codex how to call these binaries.\n\n## Why a desktop AI assistant needs LibreOffice\n\nPDF extraction, docx parsing, table reading — the model cannot do these itself, real tools must execute them. By bundling LibreOffice, Codex can convert PDF to docx, docx to PDF, or run in headless mode for batch conversion. Python and Node.js provide the runtime substrate that lets skills invoke arbitrary scripts. This is a micro AI workstation, not a chat client.\n\n## 1.7GB versus 10GB: two different paths\n\nBut the real counterpoint is Anthropic's Claude Desktop. GitHub issue #22543 has been documenting this for months: every Cowork session writes a 10GB rootfs image to `~\u002FLibrary\u002FApplication Support\u002FClaude\u002Fvm_bundles\u002Fclaudevm.bundle\u002Frootfs.img`, never cleaned up, regrows the day after deletion. On an 8GB Mac, idle CPU sits at 24%, climbing to 55% within minutes, swap steadily rising. The temporary workaround is violent: `rm -rf` three cache directories, task speed improves 75% — but only temporarily, the underlying memory leak is unfixed.\n\nOpenAI's 1.7GB and Anthropic's 10GB differ in packaging philosophy. Codex follows the \"bundle all native binaries\" route: small footprint, fast startup, depends on system libraries. Claude follows the \"embed a full Linux VM\" route: stable functional isolation, but disk and memory overhead are an order of magnitude higher. Both paths point to the same conclusion: if an Agent is to do real work locally, model weights alone are insufficient — it needs a complete traditional toolchain.\n\n## The Agent paradigm is reverse 1990s-ifying\n\nThis is the paradigm inflection behind Agent products. The earlier Copilot generation placed code generation in the cloud IDE, with only a lightweight plugin locally. The new Agent generation reads files locally, runs scripts locally, manipulates Office documents locally. Calling local tools directly is more realistic than stacking every capability into the cloud model. The local toolchain is unavoidable — when users hand the Agent real documents, real PDFs, real code repos, the Agent has to treat them as binary objects in the local environment.\n\nThe bill lands on user disk and memory. 1.7GB is not extreme — but codex-primary-runtime is a single app's cache; Claude Desktop's vm_bundles sits at a permanent 10GB. A developer running Codex, Claude Desktop, and Cursor concurrently consumes 30GB in local AI toolchains. Layer in Hugging Face caches, uv, npm, Python venvs, and a 512GB laptop fills fast. Willison's screenshot shows huggingface alone at 356GB.\n\nThe industry's next move is either to learn from Docker Desktop — image dedup and on-demand loading — or to learn from cloud IDEs — push the toolchain to remote containers, leave only a lightweight client locally. From 1.7GB to 10GB, we see the product decision; what we do not see is the Agent paradigm eating the desktop back into a 1990s-style heavy-client era where you \"install the whole toolchain.\"\n\nSo before installing the next AI client, check what it intends to stuff onto your disk.","codex-bundles-libreoffice-ai-desktop","2026-09-03T03:00:00Z","2026-09-03T09:10:19.042819Z","2026-09-03T09:10:19.042826Z",true,"agent",176,[39,48],{"slug":40,"tag_slug":40,"title_zh":41,"title_en":42,"intro_zh":43,"intro_en":44,"id":45,"is_active":35,"created_at":46,"modified_at":47},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":49,"tag_slug":49,"title_zh":50,"title_en":51,"intro_zh":52,"intro_en":53,"id":54,"is_active":35,"created_at":55,"modified_at":56},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":58},[59,64,69,74,79,84],{"id":60,"title":61,"news_slug":62,"published_at":63},"1942b07b-f794-42b1-b944-ca6b32d4ae16","四大 AI 模型同日集体掉线:OpenAI\u002FClaude 官方确认,Gemini\u002FGrok 表面沉默","four-ai-models-overlapping-outage-sept-2026","2026-09-06T08:00:00+00:00",{"id":65,"title":66,"news_slug":67,"published_at":68},"390c2437-4e4f-45ec-8270-67c5bfa4fa47","ChatGPT、Claude、Grok、Gemini 罕见同时下线,周四早晨全球 AI 集体失声","chatgpt-claude-grok-gemini-thursday-outage","2026-09-05T06:00:00+00:00",{"id":70,"title":71,"news_slug":72,"published_at":73},"9e58d587-3c1b-44c5-ad36-daf23aeb42a2","微软叫停 tokenmaxxing:GitHub Copilot 默认切回 GPT-5.6 Sol,Parikh 设 token 预算","microsoft-token-budget-gpt-5-6-default","2026-09-03T00:30:00+00:00",{"id":75,"title":76,"news_slug":77,"published_at":78},"7dec6918-b6cb-4b85-a6bf-88d1abc332d0","加密推理块漏洞让 Anthropic\u002FOpenAI\u002FGoogle 的思维链全部裸奔","stealing-reasoning-traces-llm-apis","2026-08-21T10:00:00+00:00",{"id":80,"title":81,"news_slug":82,"published_at":83},"6e79fd96-2b0f-4743-b7ac-6b39f875f2cb","AISI 122 轮 cyber eval 图解：17 次 Mythos 5、2 次 GPT-5.6 Sol 越界","aisi-cyber-eval-mythos-gpt56-august-2026-deep-dive","2026-08-09T02:00:00+00:00",{"id":85,"title":86,"news_slug":87,"published_at":88},"3967306f-062a-41a6-ab58-f99e70fc0e68","AISI 122 轮 cyber eval 越界：OpenAI 与 Anthropic 同日披露","aisi-mythos-5-gpt-5-6-cyber-eval-incident-2026","2026-08-08T04:00:00+00:00"]