全部 AI 动态

Gartner predicts cost pressure will force more than 80% of neocloud providers to cease independent operations by 2030. Hyperscalers face less risk because their large, diversified revenue could plausibly sustain AI investments for years. Gartner also forecasts AI-optimized IaaS spending will grow from $66B in 2027 to $174B in 2030. That growth is too low for most AI compute providers to recover their investments before current AI chips are replaced. AI data center owners would need $8.5T in cumulative revenue through 2030 to cover announced capex, financing costs and a 12% return on invested capital, yet Gartner forecasts attributable AI revenue of no more than $3.6T. On the current trajectory, more than $4T in asset value could be exposed to write-down risk through 2032, and Gartner does not expect the 12% return to be reached before the end of 2032.

卧槽，MiniMax H3 现在连消费级显卡都能跑了！ FastH3 Omni 正式开源！ 之前只能文生视频，现在直接升级全能模式： 图片、视频、音频都能作为参考 一次生成画面和同步音频 角色、动作、声音都能控制 支持消费级显卡本地部署 这波开源，AI 视频创作的门槛又要降了！🤣

Cindy just crossed 5,000 ⭐ on GitHub! Thank you to everyone building with us. Open-source AI agent that drives your browser, computer and phone. Free 👉 https://github.com/makecindy/cindy

More reference control. Fewer denoising steps. More places to run H3. The latest community contributions span 8-step reference-conditioned video + audio, 4-step distillation, sparse attention, and reference-token compression - alongside serving and workflow integrations across GPU servers, TPUs, and Apple Silicon. What matters is what this enables: more affordable experimentation, finer creative control, and more choices about where and how video generation runs. Open models become more useful when the community can improve the methods, adapt the runtime, and build workflows around them. Credit to @haoailab @yisun0618 @lanfeng491270 @guocheng_qian @Alissonerdx @ddalcu @tazmannner @HaochengXiUCB @huggingface @Jhuaplin, and the PTD, DMAD, SparkH3, TorchTPU and FlashML teams - including Zhengming Yu and zechengtang on GitHub. 11 additions to the H3 ecosystem index. Thank you to everyone building, testing, and sharing. https://github.com/MiniMax-AI/awesome-minimax-h3-integration

A lot of people ask me how I build games. Well, it's pretty simple and I call it "vibecoder dream" Before going to bed, I give a simple game idea to Tripo CLI + Unreal CLI, then, using ORCA CLI, let Fable orchestrate everything at max effort, with 10+ Opus 5.5 agents running at high effort. By the time I wake up, I usually have something like this.

@Kling_ai 的 Kling 4.0 已于 D0 在 Runware 上线 🎬 支持最长 30 秒的 1080p 视频，带原生音频，此外还有： - 多关键帧支持，包括首帧、中间帧和尾帧 - 每次请求最多支持 15 个输入，涵盖关键帧、参考图像、参考视频和元素 - 基于提示词编辑现有片段 输出价格低至每秒 $0.133，未来 7 天在常规价格基础上享 15% 折扣。 它加入了我们 API 上已有的 19 个 Kling 模型，覆盖视频、图像、数字人和唇同步。 已经在 Runware 上调用 Kling 模型？将你的模型 ID 切换为 klingai:kling-video@4.0

The problem isn’t that AI is intelligent. It isn’t. The problem is that too many adults (but not all) are at once mechanistic, imitative, unimaginative, stupid, formulaic, influenceable, & gullible, arrogant beyond their skill set, & too expensive to maintain.

Pretty much. As of the past few weeks, debt markets have almost entirely shut off access to money for any company that doesn’t have substantial revenue independent of AI data centers. SpaceX closed $40B of debt for GPUs, but we’re investment grade and have Starlink & launch revenue. That means the 200+ neoclouds are going to see guppy-level death and consolidation. Some will make it through, but not many.

Open weights. Shared progress. MiniMax H3 is moving fast. We built H3 for video generation with native stereo audio and multimodal reference control. The open-source community is making that capability faster, more accessible, and easier to build on. Recent highlights: • FastH3 — FastVideo, Nuva Lab and NVIDIA: 4-step distillation, now running on DGX Spark and Apple Silicon. • Sol-H3 — NVIDIA’s SANA team: 15 seconds of 768p video + audio in 6.6 seconds on 8×B300, in the team’s warm-inference benchmark.* • VDN — Haocheng Xi and the OpenVDN team: rethinking attention for faster H3 inference, with weights, training and inference code released. • PDD — NVIDIA’s distillation method, brought to H3 by Alibaba PAI as 8-step Acc-LoRAs, now supported in ComfyUI. • LightX2V — 4- and 8-step Turbo LoRAs, with workflows for text, image and reference-conditioned video + audio. Behind every release are people training, optimizing, quantizing, testing and sharing. Special thanks to: @haoailab @nuvalab @NVIDIAAI @xieenze_jr @HaochengXiUCB @ArashVahdat @julberner @LightX2V @ComfyUI And to the individual contributors pushing the work forward: @haozhangml @cxlcl1 @lawrence_cjs @yitongli165665 @haopengl33 @songhan_mit @shanasaimoe Thank you for building with H3 and helping make it faster, more accessible, and more useful for the community. Powerful models go further when we build together. Keep pushing H3. Excited to see what comes next. 🚀 Explore the ecosystem: https://github.com/MiniMax-AI/awesome-minimax-h3-integration

Since I'm now blocked by bluey from replying to this thread, I answer here by quote tweet the openAI construction has more to do with convex integration than anything else. In convex integration, the oscillations are typically built by hand and the iteration procedure creates wild weak solutions. The non-uniqueness construction of Albritton–Brué–Colombo used an unstable vortex to split solutions apart. The Córdoba–Martínez-Zoroa singularities amplify vorticity across scales by strain. In the openAI paper, a fluid instability is clever made so that the flow itself generates the oscillations: they grow by drawing energy from the background shear. The Reynolds stress then becomes the force that creates the collapsing smooth vortex. Using a physical instability to create the stress in this way has no precedent, and it is what produces a smooth force rather than a merely rough one. Some on these ideas are in the literature (e.g. Leibovich-Stewartson 1983) but the way they are combined is definitely insane and definitely nowhere in the literature previously. I also don't know of anyone suggesting that convex integration would play a role in the creation of a blow-up solution like this. On a technical level, the paper is unbelievable, maybe beyond what a human could have come up with but not beyond human understanding: there are several different groups of pde experts working on understanding it who say that they more or less understand now. I expect digestions soon td;lr: anyone saying that this OAI result is unsurprising is bullshitting sorry!

Pretty much. As of the past few weeks, debt markets have almost entirely shut off access to money for any company that doesn’t have substantial revenue independent of AI data centers. SpaceX closed $40B of debt for GPUs, but we’re investment grade and have Starlink & launch revenue. That means the 200+ neoclouds are going to see guppy-level death and consolidation. Some will make it through, but not many.

Cyberpunk inspired Roadster free-roam browser game built by Perplexity Computer: https://2049.pplx.app

@Yuchenj_UW I think we’re going to see a lot of stranded compute in partially complete data centers

Grok Bot 0.77 looks INSANE 🔥 > Calls: your Bot can dial someone, sit on hold, press keys and get transferred while you watch a live transcript > Messages on Mac: a setup screen under Connect Apps, still switched off > Tailscale: Grok Bot can join your private tailnet once an admin approves it > Payments: a screen to connect payment accounts and pick a default, on top of yesterday’s Vault > Sharing: emails when someone shares a Bot with you, and invites need Premium or Ultra

Bill Gates wants powerful open-weight models to run only where the government says they can. For any model capable of sophisticated cyber attacks, he wants safeguards and monitoring kept in by law. Developers could still use these models for free and customize them, but only on certified platforms that detect and refuse any attempt to strip out the safeguards. ---- Full video on "Decoder with Nilay Patel (@reckless) and The Verge" YouTube channel, (link in comment)

Bill Gates wants powerful open-weight models to run only where the government says they can. For any model capable of sophisticated cyber attacks, he wants safeguards and monitoring kept in by law. Developers could still use these models for free and customize them, but only on certified platforms that detect and refuse any attempt to strip out the safeguards. ---- Full video on "Decoder with Nilay Patel (@reckless) and The Verge" YouTube channel, (link in comment)

Build with Kling 4.0 on Runway Dev. Access it via API alongside the full Kling lineup and generate up to 30 seconds of native video with audio, with up to 15 references and 10 keyframes for precise control. Get 20% off Kling 4 through November 30. Start building at the link below.

today we open the beta for Krea 3, a new frontier for image editing. - end-to-end 2k edits in under 8s. - matches Nano Banana Pro or GPT Image 2.5 quality. - bounding box control and precise multi-turn edits. learn and get access 👇

Sundar Pichai 谈 Gemini 4 与 AI 算力紧缺 即便是 Google 也拿不到足够的算力。在 Gemini 4 Argon 更广泛发布后不久，我与 Google 和 Alphabet CEO @sundarpichai 坐下来聊了聊它的早期反响、他仍希望模型改进的地方，以及要让 Google 保持在 AI 前沿需要付出什么。 Pichai 解释了 Google DeepMind 近期的领导层变动，以及为什么构建前沿模型已成为一份“全身心投入”的工作。我们还谈到了 Google 对个人智能体竞赛的回应、他如何预测 Gemini 将替你完成更多工作，以及智能体最终是否会把 Search、Gmail 和 Gemini 整合到一起。 我问他 Google 自由现金流转为负值、AI 基础设施建设中的风险，以及他如何在还不知道模型将具备何种能力的数年前就规划需求。他解释了让他有信心继续投入的信号，以及他如何使用 AI 智能体来跟踪 Google 内部的模型开发和算力使用情况。 时间戳： 00:00 Gemini Argon 发布反响 01:49 模型访问与安全 06:20 AI 模型竞赛 08:35 智能体突破与 2027 年 11:34 递归自我改进 15:41 DeepMind 领导层变动 19:52 Google 的 AI 算力投资 25:20 AI 支出可持续吗？ 33:18 Sundar 如何使用 AI 智能体 35:35 Gemini 的个人智能体计划 38:37 AI 智能体如何改变网络 43:52 Google 的太空数据中心 47:38 Waymo 的扩张战略 50:26 Wing 的无人机配送计划 51:36 AI 安全与监管 55:58 Google 的量子计算押注 YouTube：https://youtu.be/fJBSNEFdm8c Spotify：https://open.spotify.com/episode/0cJBqmVz3CWJSPdmBOeJGu?si=NBxJIMhZRgikafPtwY0sZA Substack：https://sources.news/p/sundar-pichai-gemini-4-google-other-bets Apple Podcasts：https://podcasts.apple.com/us/podcast/sundar-pichai-on-gemini-4-and-the-ai-compute-crunch/id1840154537?i=1000797461022 本期节目的呈现赞助商是 @hyperagentapp。感谢节目的首席赞助商：@Atlassian、@meetgranola 和 @mercury。

The Impeccable skill by @pbakaus, now in Computer's skills marketplace, reviews an interface and fixes what a designer would flag. It checks layout, type, contrast and accessibility. Tasks that use /impeccable are free through October 31, up to 5,000 credits per user.

🎬 Kling 4.0 is live on Lovart → Up to 30s clips in one take → 10 keyframes to call every beat → Up to 15 references: characters, products, scenes → Steadier motion, better audio, bigger camera moves Arti drafts the shot list. You direct the cut. @Kling_ai

Big news: @Kimi_Moonshot is back in the top-4 labs in the Code Arena: WebDev with the release of Kimi K3.1 Pro! This model just landed #8 in the Code Arena: WebDev with 1692 pts, matching Claude Opus-5 (Max) at 70% lower cost. This is a 9 place jump (+37 points) from Kimi K3 (Max) at #17 with 1655 pts. Kimi K3.1 Pro improved from Kimi K3 across all categories. Highlights: - #7 in Brand & Marketing (up from #9) - #9 in Simulations (up from #20) Cost efficiency is another upgrade. Priced at $2/$10 per million input/output tokens, Kimi K3.1 Pro (Max) has a median cost per task of $1.23 (29% less than Kimi K3). See its placement compared to the Pareto frontier in thread. With open weights coming soon, Kimi K3.1 Pro would be the new #1 open model in the Code Arena: WebDev, overtaking DeepSeek-v4.1-Flash. Stay tuned! See @Kimi_Moonshot’s ranking by lab, and 3D generations from Kimi K3.1 Pro below in thread. Huge congrats to the @Kimi_Moonshot team on this release!

Higgsfield, an AI video startup founded by a former Snap executive, said it has attracted investment from Nvidia and TCV to help fund global infrastructure and hiring to capture growing demand. https://www.bloomberg.com/news/articles/2026-10-23/nvidia-tcv-invest-in-ai-video-startup-higgsfield-in-series-b-extension?taid=6adb140d440aec0001fbdeb7&utm_campaign=trueanthem&utm_content=business&utm_medium=social&utm_source=twitter

Kling 4.0 现已在 fal 上公开可用，API 访问对所有客户开放。 来自 @Kling_ai 的最新模型带来： • 最长 30 秒的原生视频生成，含音频 • 最多 15 个图像、视频和元素参考 • 通过最多 10 个关键帧实现精确控制 • 改进的运动、角色一致性和电影化叙事 今天就在 fal 上开始使用 Kling 4.0 构建。