搜索“xAI”

、阶跃星辰等也在与天猫接洽中。 用户支付固定月费，就可以在一定额度内使用编程工具和模型；额度触顶后，可以等待重置、购买额外额度或者转向API付费。由此，模型竞争的战场也从“每百万Token多少钱”，转到了套餐能提供多少使用额度、实际调用是否稳定，以及能否高效完成任务。 近日，SemiAnalysis 把 OpenAI、Anthropic、Meta、SpaceXAI、MiniMax、月之暗面、智谱、Cursor 和 Cognition 的订阅套餐逐个测了一遍。它首先测出输入、缓存写入、缓存读取和输出会消耗多少订阅额度，再按自己 9 月的 Agent 使用结构计算，其中约 96.6% 是缓存读取，最后套用各家的 API 标价，看同样一批 Token 如果直接走 API，需要花多少钱。 结果显示，Claude的订阅补贴力度远超国产模型。每月200美元的Max 20x套餐，使用Sonnet 5.5最多可跑出约12529美元的API用量，相当于月费的62.6倍；Opus 5.5 约为 11726 美元，相当于 58.6 倍。 而国产Coding Plan虽然月费更低，但算下来补贴力度并不高。13

匹配字段正文

管控就越严，这是行业里第一套成体系的动态安全机制。到 2023 年 10 月，Google 领投的一轮让 Anthropic 估值站上 184 亿美元。全球大模型行业告别了早期的 “一超多强”，开始隐隐显现出两家顶尖独立实验室双雄并立的雏形。这一年全球 AI 圈也风起云涌，GPT-4 发布树立新的能力标杆，OpenAI 成立专门团队押注长期安全，马斯克创办 xAI，谷歌整合 DeepMind 推出 Gemini。国内则开启 “百模大战”，百度、阿里等大厂的大模型集中亮相，智谱、月之暗面、深度求索等创业公司也开启密集融资。2024 年是 Anthropic 的爆发年，在各个方向都拿出了代表作。3 月，Claude 3 系列一次性推出三款产品，覆盖旗舰、均衡、轻量三大场景，并补齐了多模态能力。其中旗舰模型 Opus 默认支持 100K 的处理长度，在长文本评测中表现出极高的准确率，首次反超 GPT-4，将长上下文技术转化为扎实的产品竞争力。随后在 6 月登场的 Claude 3.5 Sonnet 模型能力迎来大幅提升，在多项核心指标上追平 GPT-4o；到 10 月推出升级版时，该模型在代

匹配字段正文

公示机制想解决的：与其硬闯，不如公开邀请，让每个 App 自己选。选“允许”，豆包就能以代理身份进店；选“拒绝”，门就锁上。30 天里，没有人选“允许”。 美国的 App，真的更“好说话”吗 把镜头转到海外，画风确实不太一样。 Operator 在美国能进一大片网站下单；Muse 背靠 Meta 的庞大生态，天然就长在用户每天打开的入口里；Grok 借着 xAI 和硬件公司的绑定，也在抢“系统级助手”的位置。表面上看，美国那边的 App 公司似乎更愿意给第三方代理发门禁卡。 但这个对比要打折扣。美国也并非门户大开——大量网站用 robots.txt、反爬策略、登录风控把自动代理挡在门外，Operator 自己也在隔离环境里跑、关键操作要用户确认。银行、航空、医疗这类敏感行业的网站，更是把自动代理当异常流量拦。Operator 真正能顺畅跑下来的，是那些标准化、低风险的电商和表单类站点，远没到“随便进哪家店”的地步。真正让美国 agent 比中国同行“活得下来”的，不是大厂更慷慨，而是三件中国这边缺的东西： 一是 Web 的开放传统。美国的服务很多以网站形态存在，底层是开放的 HTTP

匹配字段正文

蒂亚诺，今年9月正式加入了OpenAI基金会董事会。制定评估标准的人，坐进了被评估公司的治理层，“既当运动员又当裁判员”的争议随之而来。 “以安全为名协同减速”是否越过了分界线，如今已不止是猜测：9月18日，四名分别使用ChatGPT、Claude、Grok、Gemini的付费用户在美国加州北区联邦地方法院提起集体诉讼，指控Anthropic、OpenAI、xAI与谷歌达成非法协议、联手放缓研发，涉嫌违反反垄断法。 另一方面，在达里奥、山姆·奥特曼、马斯克呼吁“暂停”前不久，他们各自掌舵的Anthropic、OpenAI、Grok也刚刚发布过最新模型，且大多出了降价举措，如Claude Opus 5.5单价直降20%，Anthropic在宣传中更表示典型任务实际成本降约40%；GPT-6 Sol的价格相比上一代几乎打了五折——巨头本身也在悄悄搞促销。 真相可能就藏在夹缝里：事故是真的，所以刹车是必要的；但手握刹车的，恰恰也是最想独占赛道、最擅长把“安全”变成准入门槛的那几家。真正的风险预警，必须落在对所有人一视同仁的规则里——而不是某些公司既当运动员，又当裁判，还负责给其他车手划线。

匹配字段正文

for any given task, including Claude Opus 5.5, MidJourney, Suno and other leading APIs. Whatever is most likely to give you the best outcome.” Grok Bot is a joint product of SpaceXAI, the Musk AI lab now folded into SpaceX, and Cursor, which SpaceX agreed in June to buy for $60 billion in stock, a deal that closed in August. In August, I wrote that when models converge on price, the money moves to whoever decides which model gets the job. This trend is happening across organizations increasingl

匹配字段正文

ENEMY!” The same tech company figures who rushed to donate millions to Trump after his election have quickly signed on. Trump’s “First Buddy” Elon Musk has promised to rename SpaceXAI to SpaceXSI, while Jeff Bezos declared he likes it because “artificial is not a very flattering thing.” Jensen Huang, whom Trump called live on stage last month to rant that “robots will not be taking over,” has likewise adopted the term. Sam Altman praised it too, though he said OpenAI won’t be changing its name.

匹配字段正文

nthusiasts have been using agents for a minute now — homebrew OpenClaw setups led to a surge in Mac Mini sales earlier this year. But the launch of Meta’s Muse, OpenAI’s Dots, and xAI’s Grok bot has brought easy to use agents to millions. Muse and Dots have had the highest-profile product launches, and they’re fascinating to pit against each other. Both Meta and OpenAI have decided to pitch these agents to mainstream users and businesses in the form of cute, animated mascots . These can do every

匹配字段正文

ult current, steady voltage, and black start. An island has to supply all of that itself — power systems folks know how hard this can be. AI workloads make the job even harder. At xAI’s first Memphis site, swings of 10 to 20 MW several times a second were wearing out turbine shafts until xAI added 150 MW of Tesla Megapacks. A grid ERCOT’s size dilutes swings like that, but on a private plant the turbines really feel it. Expect more batteries and other energy storage as rack-level power density g

匹配字段正文

rt hundreds of thousands of users, including Instinct, Tomo, Poke, Lindy, and Town. And, the incumbents have struck back with agents of their own — Meta’s Muse, OpenAI’s Dots, and xAI’s Grok Bot. Most of these agents strive to be available everywhere — like a human assistant, you can text, email, or call. The channel du jour is iMessage, where viral agent Instinct launched in August 2026. Companies like DoorDash have since launched their own iMessage agent for text-to-order. More prosumer-orient

匹配字段正文

g internal and external red-teaming of AI models for “severe and novel threats” and also “to work toward information sharing” on AI safety issues. Google, Anthropic, Meta, OpenAI, xAI, and Nvidia did not respond to requests for comment. The White House also did not immediately respond. Neil Chilson, a former chief technologist for the FTC, posted on X of the accords that “this is the kind of pledge that the FTC could potentially enforce, if a company materially failed to follow through on any of

匹配字段正文

respond faster to misaligned behavior.” OpenAI’s rivals have taken note. Spurred by the fallout from the incident, the major AI labs—including Anthropic, Google DeepMind, and SpaceXAI—have all called for the pace of development to slow down. But how does that square with fierce international competition and trillion-dollar IPOs? “We’re not going to shoot ourselves in the foot and take ourselves far off the frontier—that’s just a horrible strategy,” he says. “I think it’s really about setting a n

匹配字段正文

验室”（The Generation Lab）联合进行了一次调查。他们访问了1088名18至34岁的美国年轻人，并问了一个很直接的问题：你是否相信那些站在AI浪潮最前沿的企业领袖，会负责任地发展AI？ 调查列出了九位科技界的重量级人物——OpenAI的山姆·奥尔特曼（Sam Altman）、Meta的马克·扎克伯格（Mark Zuckerberg）、特斯拉和xAI的埃隆·马斯克（Elon Musk）、英伟达的黄仁勋（Jensen Huang）、谷歌的桑达尔·皮查伊（Sundar Pichai）、Anthropic的达里奥·阿莫迪（Dario Amodei）、微软的萨提亚·纳德拉（Satya Nadela），以及Palantir的亚历克斯·卡普（Alexander Karp）和彼得·蒂尔（Peter Thiel）。但没有一位得到了多数受访者的信任。最不受信任的卡普，不信任比例达到81%；即使表现最好的纳德拉，仍有65%的年轻受访者对其表示不信任。与此同时，有45%的受访者认为，AI会对自己的职业生涯产生负面影响，只有10%认为它会带来正面影响。 如果仅有这样一个调查，我们当然不能作过度解

匹配字段正文

ry has adapted in two ways: Foundries shrink the cost of doing. They industrialize the measurement, using new technology to generate data an order of magnitude faster than before. Xaira, NewLimit, Octant, Tahoe, and Endura accomplish this with next-generation sequencing and multiplexing; Insitro, Eikon, and Noetik with high-throughput microscopy; Lila and Periodic Labs with physical automation, just to name a few. AI makes that data legible and predictive, but the differentiating asset is the ex

匹配字段正文

ns remain closed while only older versions are open, and joke that Meta’s infinity logo implies an indefinite wait. Commenters contrasted Meta’s unreleased Muse/Spark weights with xAI’s Grok release pattern , noting that “Grok 4.6 (4.7 upcoming)” exists while only Grok 1 and Grok 2 have been open-released, implying a widening lag between frontier closed models and published weights. A technically relevant explanation linked to Mark Zuckerberg’s post on X: x.com/finkd/status/2099997096896274533 .

匹配字段正文

到 Qwen4 发布时，生态系统将修复集成问题。 zai-org 的 GLM-5.3-Flash：此次发布完善了我们在 2025 年写过的中国模型玩法：该模型以免费使用的“隐身模型”形式发布，名为“Ox-Alpha”，在 OpenRouter 和 OpenCode 上亮相，这首先让人们兴奋地想要尝试它。随后他们猜测其创建者和规模，声称它是来自 Cursor/xAI、Gemini 的 1T 模型，或是开源实验室的新预训练模型。由于该模型性能相对出色，人们连续数天猜测其创建者，从而积累了热度。这也削弱了伴随每次（开放）模型发布而来的“刷榜”（benchmaxxing）指责。 tencent 的 Hy4-preview：腾讯正在成为开放模型领域的重要参与者，在扩大其旗舰模型规模的同时，转动后训练飞轮。其成果 Hy4-preview 是一个有能力的模型，目前存在过度思考的问题。然而，如果从 Hy3-preview 到 Hy3 的轨迹有任何指示意义，最终模型可能会真正冲击开放模型的前列。 在我们的 Artifacts Hub 上查看本期所有模型的更多详情。 访问 artifactshub.ai

匹配字段正文

es, it created new companies (AI companies) and industries (companies to serve AI companies); $217 billion of that $400 billion went into OpenAI and Anthropic, $20 billion went to xAI and $16 billion went to Waymo, with NVIDIA investing over $40 billion itself; I’m still not sure what this was meant to communicate other than investor fluff; As for the industries it’s created, it’s unclear. Even if we were being generous to describe the companies that exist to throw a layer over an existing AI mo

匹配字段正文

ed revenue of the entire generative AI industry was around $175 billion annualized run rate as of June 2026. Every other dedicated model developer (e.g. the Chinese labs, Mistral, xAI before the SpaceX acquisition) was below $1 billion/year in run rate as of late 2025 or 2026; The growth rate within 2026 may be slowing down. For example, the two labs were reportedly around $55B combined at the end of Q1, nearly doubling in the first quarter, but took around four months to double again (still a f

匹配字段正文

tary information. The state of affairs on peak model performance is roughly as follows : Anthropic – Claude Fable 5 OpenAI – GPT 5.6 Sol Moonshot AI – Kimi K3 (open weights ) SpaceXAI – Grok 4.5 Zhipu (Z.ai) – GLM 5.2 (open weights) Meta – Muse Spark 1.1 DeepMind – Gemini Flash 3.5 Alibaba – Qwen 3.7 Max (3.8 announced, also to be open-weights, when writing) It is astonishing to see DeepMind, and some of the other American giants this low. In many ways, the X AI team deserves more credit. A visu

匹配字段正文

d GH200 GPUs, plus 1,328 installing. That is more than 11x the tracked A100 total of 41,208. The A100 chart now reads like a legacy fleet chart; Hopper is the live installed base. xAI Colossus 1 is the largest tracked Hopper deployment at 200,000 GPUs, and by itself is almost five times the entire tracked A100 set. Tesla Cortex follows at roughly 66,000 H100-equivalent GPUs, then Meta’s GenAI clusters at 49,152, a CoreWeave H200 cluster at 42,000, Voltage Park at 24,000, and Germany’s JUPITER Bo

匹配字段正文

pending an estimated $450bn a year on AI infrastructure. Its answer has been the EU AI Act and a capital commitment that is a rounding error by comparison. A single American site, xAI’s Colossus in Memphis, runs more than half a million GPUs. Europe has nothing remotely at that scale. The instinct to govern this technology is right, but we’re off on the ambition by orders of magnitude. It is fair to object that regulation is itself a form of power. But a rule book is not a substitute for the thi

匹配字段正文