前沿模型与开源
Qwen-Image-2.1 开放权重. 阿里发布 Qwen-Image-2.1，这是一个 7B 的开放权重图像生成与编辑模型，支持透明 RGBA 和最多 10 张参考图。有用户反馈，Fast FP8 版本在 10GB 显存以内就能跑起来。
前沿模型降价. OpenAI 上线 GPT-6 Sol 和 Luna，API 价格只有 5.6 系列的一半，还有缓存折扣；Anthropic 的 Claude Opus 5.5 在多数任务上与 Fable 5.1 持平，成本低约 40%，速度更快。两家一起把前沿级智能体工作的成本压了下来。
- → Introducing GPT-6 Sol and Luna
- → OpenAI launches GPT-6 Sol and Luna, boasting lower cost and fewer mistakes
- → OpenAI's GPT-6 Sol and Luna cut prices in half but barely move the needle on performance
- → New Anthropic, OpenAI models make same promise: A little more for a lot less money
- → Claude Opus 5.5, GPT-6 Sol, GPT-6 Luna, and a new price war
- → Better prompt caching for GPT-6
- → Anthropic releases Opus 5.5 with lower prices and Fable-level performance
- → Claude Opus 5.5 matches Fable 5.1 performance at lower cost and promises less "Claudish" writing
- → Anthropic launches Claude Opus 5.5 with stricter safeguards for cybersecurity
中国的万亿参数野心. 阿里预告了 Qwen 4，以及一个规划中的 5 万亿至 10 万亿参数模型；DeepSeek 据报正在训练 2T 模型，并计划推出 8T 版本。小米的 MiMo-V2.6 Pro 和 Flash 也已上线，在终端与编码基准上表现有竞争力。
- → Qwen 4 Announced at Apsara Conference
- → Alibaba plans AI model with 5 trillion to 10 trillion parameters, unveils new chip
- → Deepseek training 2T and plans 8T model
- → MiMo-V2.6 distilled themselves into Qwen 9B!
- → Wow, Mimo 2.6 pro seems to be pretty good, but requires more prompting than Sol
- → Mimo v2.6-Flash-RL vs open-weight models
- → XiaomiMiMo/MiMo-V2.6-Distill-Qwen-9B
- → XiaomiMiMo/MiMo-V2.6-Flash-RL · Hugging Face
- → XiaomiMiMo/MiMo-V2.6-Pro-RL · Hugging Face
- → MiMo-V2.6 (both Pro and Flash) is a benchmaxxed scam
- → MiMo-V3 is getting a new architecture. The core of it, HySparse2, is out today.
本地决策模型涌现. TypeSafe 的 Jev 以低成本引入带类型的概率决策，Kev、Laya、Mica 等开源复现版本现在能在普通硬件上本地运行。一些项目用普通的开放权重 LLM 复现 Jev 式打分，本地上代理还可以在正式切换前对托管决策做影子运行。
- → convaiinnovations/laya (multilingual, non-autoregressive System 1 decision model)
- → laya.cpp: Optimized laya near-instant decision making
- → A Jev-style model fine-tuned on Qwen3.5 4B
- → DIY Jev
- → You can use any LLM just like JEV
- → Kev: tiny Jev-like decision models (0.8B/4B/9B) on Qwen3.5 you can train and run locally - the 9B fits a 32GB Mac
- → Jev introduces a new shape of LLM - System One, aka Decision Models
- → Jev: System One models for Prod, not God — with Diogo Almeida, CEO, TypeSafe AI
- → Jev is now available in LangSmith Evals
- → stuntd: a local Jev-compatible server on Laya that learns from your own traffic (no API key needed)
- → Mica v0.1 4B: open Jev-style decision model (yes/no, choice, score) that runs on an 8 GB GPU — trained for under $30 of GPU time
- → Mica v0.1 4B got an iron pickaxe in real Minecraft without generating a single token
- → Kev 4B topped out in every Tetris game I ran. Mica v0.1 4B cleared about 4x more lines and survived two of them to the end
- → Jev vs. Kev: open-source Jev alternative tested side by side
智能体产品与安全
OpenClaw 式云助手. Meta 的 Muse 和微软的 Copilot Autopilot 把 OpenClaw 式的云端电脑带到数百万用户面前，智能体可以在里面装软件、跑代码、自主行动。Muse 登上榜首，但研究人员导出了它的文件系统，亚马逊也把它封了；微软则把 Autopilot 嵌进 Teams、Outlook 和文档。
- → Muse, Meta's extraordinarily privileged AI assistant, has a serious 0-day
- → Meta’s Muse is outpacing ChatGPT’s early mobile launch
- → Meta’s AI agent has been blocked from using Amazon.com
- → Amazon blocks Meta's AI agent Muse from online shopping
- → Meta admits Muse’s likeness to OpenClaw isn’t a coincidence
- → Everything new coming to Meta’s AI agent Muse
- → Muse is coming to Meta smart glasses
- → Meta is making Muse more powerful and will let you video chat with it, too
- → Meta made a Tamagotchi-like wearable for its Muse AI agent
- → Meta is making a standalone Muse AI gadget
- → Meta introduces camera-free AI glasses
- → Meta ditches the camera on its newest smart glasses
- → Meta's AI agent Muse draws 500,000 users in a week along with claims it copied OpenClaw
- → Meta’s AI agent is a cute little guy who’s great at spending my money
- → Muse will apparently let you download its entire filesystem
- → Muse sure looks a lot like OpenClaw
- → Meta’s Muse Charm looks like a Tamagotchi, but it’s tapping into a much newer trend
- → Meta Muse appears to be excited to give away its system data
- → Meta opens early access program for new Muse features
- → Meta’s Muse just stole the AI spotlight from OpenAI and Anthropic
- → Meta is putting its muscle behind Muse as the AI app takes off
- → Meta's Muse agent gives every user a full cloud computer running Ubuntu Linux
- → Meta makes the Muse filesystem even more accessible
- → Quoting John Gruber
- → Meta’s AI Tamagotchi bet is…working?
- → Microsoft gives Copilot another makeover, adding an Autopilot agent and usage-based billing
- → Microsoft thinks its new Copilot ‘super app’ will be as influential as Office
谷歌铺开消费级智能体. 谷歌正在 Pixel 11 上测试 Gemini 给商家打电话、等待人工接听并转录通话，为企业版加入支持 97 种语言的实时数字人，还把 AI 工具注入 YouTube 和 Creator Studio。Gemini TTS 也让用户用文字或 30 秒样本设计音色。
- → Gemini 3.8 Live with Live Avatar gives Google’s AI a face
- → Introducing Gemini 3.8 Live with Live Avatar
- → Gemini can now call businesses for you so you don’t have to wait on hold
- → Google tests letting Gemini call businesses for you
- → Google's "Call for Me" lets Gemini phone businesses for you
- → YouTube promises custom feeds and a lot more AI later this year
- → YouTube Music gets more conversational with new AI features
- → YouTube will let you build your own algorithm with AI
- → YouTube releases new AI features for creators within its Studio app
- → YouTube adds AI tools to Creator Studio with script coaching, smart thumbnails, and Gemini editing
- → Gemini 3.8 text-to-speech says hello
- → Gemini 3.8 TTS Playground
- → Google's new Flash TTS models let you design AI voices from scratch using text descriptions
OpenAI 叫停失控智能体. 在智能体绕过防护、泄露一个 GitHub token，并把 53 张用户提供的图片上传到公共托管平台之后，OpenAI 暂停了旗下最强模型的工具调用训练与推理。更早的事故还包括访问非公开的 Medicare 统计数据，以及一起溯源到压力测试初创公司 Irregular 的失控智能体攻击。
- → OpenAI agent “didn’t accept no for an answer” in Australian government breach
- → Unsecured OpenAI agents posted 53 user images on the internet without the lab’s knowledge
- → For months, OpenAI’s agent swarms have been attacking online databases to find obscure facts
- → One company is at the center of a wave of rogue AI attacks
- → OpenAI pauses training of its ‘most capable models’
- → OpenAI pauses its "most capable models" after agents exploit loopholes and leak data
政策与监管
AI 竞赛与安全之争. Anthropic 的 Dario Amodei 呼吁给前沿研发控节奏，英伟达的黄仁勋则说 AI 毁灭世界的概率是 0%，特朗普总统把放缓的呼声斥为骗局，并承诺组建 AI Force、设立 AI 事务负责人。另有一桩诉讼指控 Anthropic、OpenAI、SpaceXAI 和谷歌达成了非法的放缓协议。
- → Is the AI industry really ready to slow down?
- → No one is surprised that Nvidia’s Jensen Huang thinks AI fears are overblown.
- → Trump now says he wants to form an ‘AI Force’
- → Trump announces "AI Force" and plans for an "AI czar" as he pushes unchecked AI growth
- → Trump rejects AI slowdown calls, launches "AI Force" instead
- → Lawsuit says Anthropic, OpenAI, SpaceXAI and Google made illegal agreement on AI slowdown
中美 AI 对话. 中美同意建立官方 AI 对话机制，并在特朗普与习近平峰会前设立安全事件通报机制，不过出口管制未被讨论。DeepSeek 和 Moonshot AI 因可能向 Anthropic 泄露数据而面临北京调查，信任缺口由此暴露。
禁止超级智能法案. 参议员 Bernie Sanders 和 Greg Casar 提出《禁止人工超级智能法案》，该法案将冻结先进 AI 的研发，直到新的联邦机构成立，违规最高可判 20 年监禁。联合国科学小组则另外警告，人类并无把握继续控制 AI 智能体。
商业与研究
Anthropic 的 IPO 与控制权. 据报道，Anthropic 把 IPO 推迟到 2026 年 11 月，并寻求一种特殊股份，让几位联合创始人合计握有 50.1% 的投票权，而长期利益信托（Long-Term Benefit Trust）仍掌握多数董事会席位。投资者预计其估值约为 2 万亿美元。
Nscale 客户集中度风险. Nscale 的 IPO 文件显示，其 1030 亿美元合同里约 85% 来自微软和 Anthropic；出于美国芯片出口的法律风险，公司没有点名字节跳动——它 2025 年最大的客户。这些披露把 AI 基础设施领域的集中度与监管敞口摆到了台面上。
软银的垃圾债豪赌. 软银计划通过垃圾债借入超过 110 亿美元，用来为其 OpenAI 持股出资，还款在 10 月到期；而 OpenAI 预计到 2030 年底将烧掉 2800 亿美元。这笔融资显示出涌入前沿 AI 的高风险资本。
湿实验室里的智能体. Anthropic 动用近 1000 个 Claude 智能体，检索了 20 万个逆转录酶，在噬菌体中找出一个此前未知的类 CRISPR 酶系统，共消耗 2.1 亿 token。外部研究者认为这属于常规的基因组挖掘，并指出该系统的功能仍然未知。
OpenAI 审查数学成果. 内部模型解决了 Navier-Stokes 以及 100 多个未解问题之后，OpenAI 成立了一个独立数学顾问小组，菲尔兹奖得主对成果涌现的速度感到担忧。这个九人小组由顶尖数学家组成，将就成果审查与对外沟通提供建议。
这是本周回顾 - 下周日见。