全部 AI 动态

🚀 vLLM v0.32.0 is out! 814 commits · 363 contributors · 121 first-timers 🧭 New /v1/systemone structured-decision endpoint + generic hf tool/reasoning parser 🗜️ Low-precision KV everywhere: NVFP4 on SM8x/90/12x, FP8 KV default for GLM-5.x, 4-bit KV on ROCm 🛠️ Model Runner V2: GPU n-gram speculator, Elastic EP, PP+CP for MLA, now default on ROCm ⚡ Faster startup & serving: precompiled kernel downloads, autotune preload, +43–85% req/s with multi-API workers 🆕 EmbeddingGemma 2, MiniCPM-V 4.7, Nemotron 3.5 ASR 🔒 3 security fixes + CORS hardening Full notes → http://github.com/vllm-project/vllm/releases/tag/v0.32.0

Two important changes to timeline recommendations: - New AI/SI ranking model that predicts probable time spent much better - Retrieval model trained to surface posts people read all the way through User active seconds is growing exponentially and hitting all-time highs almost every day:

Run @AntLingAGI speech models locally 🎙️ Ming-3.0 is 5 open models that talk, listen and write down what you say Its voice model now ranks #1 among open models on Artificial Analysis starting at just 0.5B Run AI models locally -> https://atomic.chat

Day-0 support for Ming-3.0 in vLLM-Omni @AntLingAGI just open-sourced 5 speech models spanning TTS, audio understanding, and ASR — ranging from 0.5B to 126B total parameters (6.5B active). Serve them today with vLLM-Omni. Congrats to the @AntLingAGI team!

Ant Group just dropped its new speech and audio models! https://huggingface.co/collections/inclusionAI/ming-30

中国的互联网从13年的移动端时代就变成了“不联网”，其实现在美国的互联网也因为Agent能力重新可能变成“不联网”。 中国移动互联网 “不”联网的特征是： - 不能发现内容，搜索引擎找不到，必须进入某个 App 搜； - 不能读取内容，有链接也看不了，折叠全文、弹出登录或下载要求； - 不能使用功能，网页只有介绍或残缺功能，完整操作必须进 App／小程序； - 不能跨平台流动，链接被拦、不能直接跳转、只能复制口令或换浏览器。 这一个现象一直都很恶心，很独特的你封杀我我封杀你，平台之间的“截流”和投流量成为了一门学问，形成了我只在中国产品里见到的靠口令红包和自动读取剪贴板跳转的生态，贼tmd sb。 这也是为什么中国不管是Personal Agent还是Work Agent都是在同一体系内很能打，但是一旦跨体系就几乎智障的原因，很难出现一个Open Web上的Super Agent，都关着门，没有数据可以拿到没有功能可以调用，哪里能构建好Context和Workflow。 中国当前App “不联网” 的本质原因有两个。 一是因为2013年后手机端成为了主要入口，每一个大厂都在打造自己的超级App，于是乎web端功能不再迭代甚至删减限制，每一产品都在把聊天、内容、购物、支付乃至部分公共服务全部塞到自己生态里，再通过小程序继续确保用户不离开自己。 另外一个就是17年基本所有主流平台受网络安全法要求统一强制实名要求，其实也给了大厂们一个理由更封闭。 但是其实美国也在AI时代互联网面临着“不”联网的处境，最近大量讨论都围绕在平台是否应该向Agent开放权限上，大部分的头部传统企业也都先选择把门关上。 购物网站： - Amazon：2025年10月31日向Perplexity发函，要求其购物代理Comet停止未经Amazon许可的代理访问；2026年9月，又封锁了Meta Muse对Amazon的访问。 - eBay：用户协议直接点名代买Agent、LLM驱动的机器人等自动化工具，要求事先获得eBay明确许可才能访问其服务。 - Target：采用Agent白名单。只有用户批准、同时获得Target授权的Agent才能接入，其他Agent不能登录、查询、抓取数据，或者通过API与网站交互。 银行和券商： - 花旗Citi：不允许第三方AI Agent代客户登录账户，或者直接在其数字银行平台内执行交易；保留通过部分开放银行服务商、经客户授权的有限只读访问。 - 摩根大通Chase：要求Agent明确表明身份、标记流量，并在条款中保留限制、暂停、终止Agent访问的权利。 出行和本地生活： - Delta航空：条款明确写着，用户授权某个第三方Agent，并不构成Delta对它的授权；Delta保留拒绝与其交互、处理其请求或交易的权利。 - Yelp：未经许可的自动化访问、抓取和检索受到限制；它也提供自己的MCP接入，但需要通过Yelp的API密钥使用。

Grok Bot is now one step closer. Set it as your iPhone's Action Button and press it to talk with Bot.

MiniMax M3.1 Flash (aka Space Bunny) is now live and FREE in Command Code for limited time! · 1M token context window · Frontier level coding performance · Available all plans, all subscribers

MiniMax M3.1 Flash, aka Space Bunny, is now available on GMI Built for coding, agentic workflows, and computer use 70% lower agentic task costs, 73.7% on DeepSWE v1.1, 82% on CyberGym Run it now 👇

Today, Dream Labs is coming out of stealth with Dream-1, a robot foundation model that controls bimanual robots it has never seen, on tasks it was never shown, in places it has never been, from $3/hour. Dream API is live for everyone, today. 🧵

stealth 模型揭晓 🐰 Space Bunny 连续 10 天成为 OpenRouter 排名第一的模型，处理了 56.7T tokens！是我们运行过的 stealth 模型中处理量最多的。 Space Bunny 于 10 月 5 日离开 OpenRouter。今天它作为 @MiniMax_AI 的 M3.1 Flash 回归，并在接下来的 72 小时内免费。

MiniMax M3.1 Flash goes toe-to-toe with Opus and GPT on coding at ~25x lower cost. This was the model behind Space Bunny, the stealth model that’s been topping usage leaderboards recently. Free in Cline now!

MiniMax M3.1 Flash is on Ollama! 🔒 Hosted in the US and with zero data retention. Fast coding and agentic workflows, with image understanding and a 1M-token context window. Try it on Ollama Cloud: ollama run minimax-m3.1-flash:cloud Claude Code: ollama launch claude --model minimax-m3.1-flash:cloud Codex: ollama launch codex -–model minimax-m3.1-flash:cloud

Introducing SKY-001, the first Engineering World Model. On CAD and BIM tasks, it beats GPT-6 Astra and every frontier model we tested: -> 51-125x faster - > 150-210x cheaper -> Higher IOU than all frontier models SKY-001 lets you dream a physical object before you build it. 18 months of work. Built by the team out of Maluuba's deep learning lab (acq. by @MSRResearch). The first ‘AI General Engineer’ will be the backbone for Autonomous Manufacturing.

I used Gemini 4 Argon to make Subway Surfers for the new flying Tesla Roadster 🚀 Fly over moving traffic, hit the thrusters to barrel roll in the sky, and collect air power-ups to set a high score. Play + code below!