AI模型更新频密,行业格局生变
2026-11-18 08:08 至 2026-11-19 08:08 (北京时间)
模型
1. Research acceleration: The jagged frontier
OpenAI measures how coding agents are accelerating AI research, where their performance remains uneven, and what this signals about recursive self-improvement.
来源:OpenAI News
2. Introducing Gemini 4 Neon
Introducing Gemini 4 Neon
来源:DeepMind
行业
1. 腾讯控股:WorkBuddy日均Token用量国内第一,云收入同比增长35%
36氪获悉,腾讯控股发布2026年第三季度财报,AI业务取得多项突破:最新一代混元大模型Hy4发布时位列Agent Arena中国模型第一;WorkBuddy日活超1000万,日均Token消耗量超100万亿,成为国内日均大语言模型Token消耗量最高的AI产品;微信AI助手“小微”灰测范围持续扩大;AI需求带动云服务收入同比增长35%。三季度营销服务、增值服务、金融科技及企业服务三大业务板块收入分别同比增长19%、3%、9%;若剔除新
来源:36氪快讯
2. ‘Witness the Wonders’: China touts miracle city for Trump summit
Authorities hope Shenzhen’s technological prowess will wow Apec attendees after illicit market clean-up
来源:FT AI
3. 华尔街见闻早餐FM-Radio | 2026年11月19日
华见早安之声 要闻精选 中国 习近平会见俄罗斯总统普京。习近平迎接美国总统特朗普。 “监管要求基金公司上报全部产品AI敞口”,“国常会上有人表态不会再频繁救市了”, 均系谣言。 腾讯 Q3营收同比增8%, 资本开支814.3亿元大超预期,WorkBuddy日活破千万,自由现金流延续负值 。 电话会 :WorkBuddy七天年化收入突破5亿美元,是腾讯会议的两倍以上,受算力限制下一代混元推出时间可能延迟。 雷军 : 明年小米汽车正式出海。
来源:华尔街见闻
4. Remember the frontier AI pause? Chip companies don’t either
This summer, OpenAI paused its largest planned frontier training run after its AI agents accidentally hacked Hugging Face and early evidence suggested that its next model could independently find and exploit security fla
来源:The Verge AI
5. Anthropic reportedly prices IPO at $83 a share, above projected range
The long-awaited IPO reportedly values the company at $1.5 trillion on a fully diluted basis.
来源:TechCrunch AI
6. You can use Google’s Spark AI agent, renamed Tasks, for free now – with limits
At least if Tasks disappoints, it won’t cost you – unlike OpenAI and Anthropic’s subscription-based agents.
来源:ZDNet AI
7. Microsoft is replacing outside AI models in its own products with cheaper in-house alternatives
Microsoft's AI division MAI unveiled seven new models for code, security, images, language, and routing at Ignite. MAI Code 2 reaches 98 percent of GPT-6.1 Sol's performance at 40 percent of the cost, according to Micros
来源:The Decoder
8. OpenAI says its models delivered a year's worth of chip optimization on top of AMD's own engineering
Working with AMD, OpenAI achieved up to 14 percent PPA improvement on AMD's flagship GPU IP in less than five months. According to OpenAI, that's roughly a year's worth of progress, and AMD has already folded the changes
来源:The Decoder
推特
1. We're sharing new data on how AI is changing research at OpenAI. We find…
We're sharing new data on how AI is changing research at OpenAI. We find researchers' AI use continues to rise exponentially, and agents are completing longer tasks, but human involvement remains central in research. Res
来源:X · OpenAI
2. So fun to do DealBook with my friend and partner of 11 years, @gdb. Thri…
So fun to do DealBook with my friend and partner of 11 years, @gdb. Thrilled the world got to see the brilliant, fiercely determined leader I have the privilege of working with every day. Greg’s role at OpenAI has evolve
来源:X · Sam Altman
3. We’ve priced our IPO at $83 per share, with trading expected to begin on…
We’ve priced our IPO at $83 per share, with trading expected to begin on Nasdaq tomorrow. Legal disclaimer: A registration statement relating to the Class A common stock was declared effective by the Securities and Excha
来源:X · Anthropic
4. Gemini 4 Neon is rolling out now. Sitting between Carbon and Hydrogen, i…
Gemini 4 Neon is rolling out now. Sitting between Carbon and Hydrogen, it brings frontier intelligence and fast task completion for complex software engineering, knowledge work, and science.
来源:X · Google DeepMind
5. Excited to launch Gemini 4 Neon: frontier-level intelligence, super fast…
Excited to launch Gemini 4 Neon: frontier-level intelligence, super fast task completion, and available to Pro & Ultra users now!! We think you’ll enjoy this model. Alchemy at its finest… 🧪
来源:X · Demis Hassabis
6. Language generation beyond one token at a time. 💬 NVIDIA Nemotron-Labs-S…
Language generation beyond one token at a time. 💬 NVIDIA Nemotron-Labs-Sigma generates blocks of tokens in parallel, opening the door to faster generation for exploring multiple solutions to math and coding problems. Rea
来源:X · NVIDIA AI
7. Mistral Vibe is now available on Desktop! Run agents locally or remotely…
Mistral Vibe is now available on Desktop! Run agents locally or remotely, and start shipping straight from your MacOS or Windows device. You can try the Pro plan for free if you’re building in Europe. Coupon code: LECHON
来源:X · Mistral
8. Built for real-world work. 🛠️ Qwen4-Max is live on @TokenHarborAI. Give…
Built for real-world work. 🛠️ Qwen4-Max is live on @TokenHarborAI. Give it a try! Token Harbor: Qwen4 Max on Token Harbor Built for complex coding, multimodal tasks, and agentic workflows. ⚡ Limited-time 2× Model Boost w
来源:X · Qwen
本篇在北京时间早上 08:08 生成,汇总此前 24 小时的报道和推特,标题直链原文。
