跳到主内容
@wquguru
精选88Hacker News Best(web_list)产品发布/更新多源精选 ×5

Google DeepMind Genie 4 向 AI Ultra 用户开放

Genie 4 is now available

原文
发到 X
推荐理由

World Model 领域的重要产品化进展,Genie 4 实现了实时文本编辑与原生音频的闭环,Agent 辅助创作降低了门槛,值得关注其落地能力。

Genie 4 is now available

Genie 4 现已上线

Nov 18, 2026

2026 年 11 月 18 日

  • x.com
  • Facebook
  • LinkedIn
  • Mail
  • Copy link
  • x.com
  • Facebook
  • LinkedIn
  • 邮件
  • 复制链接

Explore a new Project Genie experience for Google AI Ultra subscribers, featuring Genie 4, the Project Genie agent, and a community gallery.

探索面向 Google AI Ultra 订阅用户的全新 Project Genie 体验,包含 Genie 4、Project Genie 智能体以及社区画廊。

Diego Rivas

Group Product Manager, Google DeepMind

Google DeepMind 集团产品经理

Pavan Kumar

Group Product Manager, Google Labs

Google Labs 集团产品经理

Share

分享

  • x.com
  • Facebook
  • LinkedIn
  • Mail
  • Copy link
  • x.com
  • Facebook
  • LinkedIn
  • 邮件
  • 复制链接

Read AI-generated summary

阅读 AI 生成的摘要

We are rolling out Genie 4 to eligible Google AI Ultra subscribers, offering more responsive, social, and interactive world generation. You can now use natural language to edit environments in real time, add native audio, and share your creations in our new community gallery. Visit the Project Genie website to start building your own worlds or join the waitlist if you are a Google AI Pro member.

我们正在向符合条件的 Google AI Ultra 订阅用户推出 Genie 4,提供更响应迅速、更具社交性和互动性的世界生成能力。您现在可以使用自然语言实时编辑环境、添加原生音频,并在我们的新社区画廊中分享您的创作。访问 Project Genie 网站开始构建您自己的世界,如果您是 Google AI Pro 成员,请加入候补名单。

Summaries were generated by Google AI. Generative AI is experimental.

摘要由 Google AI 生成。生成式 AI 处于实验阶段。

  • We are rolling out Genie 4 to eligible Google AI Ultra subscribers for creating interactive worlds.
  • You can now modify environments in real time using text prompts and native audio features.
  • Our updated Genie 3 model supports longer sessions, panorama grounding, and linked world sequences today.
  • The new Project Genie agent helps you build custom worlds through simple natural language conversations.
  • You can publish your creations to our community gallery for others to explore and remix.
  • 我们正在向符合条件的 Google AI Ultra 订阅用户推出 Genie 4,用于创建交互式世界。
  • 您现在可以使用文本提示和原生音频功能实时修改环境。
  • 我们更新的 Genie 3 模型今天支持更长的会话、全景锚定以及链接的世界序列。
  • 新的 Project Genie 智能体通过简单的自然语言对话,帮助你构建自定义世界。
  • 你可以将你的创作发布到我们的社区画廊,供他人探索和二次创作。

Summaries were generated by Google AI. Generative AI is experimental.

摘要由 Google AI 生成。生成式 AI 处于实验阶段。

We are updating Project Genie to help you create and explore interactive digital worlds. With our new Genie 4 model, you can change weather, add characters, and hear adaptive audio in real time just by typing prompts. You can also use our new creative agent to build complex environments through simple conversation. If you subscribe to Google AI Ultra, you can start building, sharing, and remixing these custom worlds in our community gallery today.

我们正在更新 Project Genie,以帮助你创建和探索交互式数字世界。借助全新的 Genie 4 模型,你只需输入提示词,即可实时改变天气、添加角色并聆听自适应音频。你还可以使用我们全新的创意智能体,通过简单的对话构建复杂的环境。如果你订阅了 Google AI Ultra,今天就可以在我们的社区画廊中开始构建、分享和二次创作这些自定义世界。

Summaries were generated by Google AI. Generative AI is experimental.

摘要由 Google AI 生成。生成式 AI 处于实验阶段。

Explore other styles:

探索其他风格:

  • General summary
  • Bullet points
  • Basic explainer
  • 通用摘要
  • 要点列表
  • 基础说明

Earlier this year we introduced Project Genie, an experimental research prototype that lets you generate and explore interactive worlds. Early testers told us they wanted deeper control over what happens inside those worlds, and easier ways to build and share them.

今年早些时候,我们推出了 Project Genie,这是一个实验性研究原型,允许你生成和探索交互式世界。早期测试者告诉我们,他们希望对这些世界内部发生的事情拥有更深的控制权,以及更便捷的方式来构建和分享它们。

So today, we are rolling out a new Project Genie experience to eligible Google AI Ultra subscribers, making these worlds far more responsive, creative, and social. This update unlocks first-time access to Genie 4, new upgrades to Genie 3, the Project Genie agent which helps users build their vision, and a community gallery to discover, explore, and remix worlds.

因此,今天我们向符合条件的 Google AI Ultra 订阅用户推出全新的 Project Genie 体验,使这些世界具有更高的响应速度、创造力和社交性。此次更新首次开放了对 Genie 4 的访问权限,对 Genie 3 进行了新升级,推出了帮助用户实现其愿景的 Project Genie 智能体,以及一个用于发现、探索和二次创作世界的社区画廊。

Project Genie brings together our most advanced world models alongside our Gemini models, empowering you to generate, step inside, and shape worlds directly from your browser.

Project Genie 将我们最先进的世界模型与 Gemini 模型相结合,赋能你直接在浏览器中生成、进入并塑造世界。

Experience Genie 4 in Project Genie

在 Project Genie 中体验 Genie 4

With Genie 4, our most realistic and interactive world model yet, you can now prompt mid-session to change the world you created. Just type a new prompt while you navigate your world and you can edit your environments in real time. Shift the weather, drop in a character, or redirect the action. For the first time, Genie 4 pairs these visuals with real-time native audio that adapts as the scene unfolds, from ambient soundscapes to character dialogue.

凭借迄今为止最逼真、最具交互性的世界模型 Genie 4,你现在可以在会话过程中通过提示词来改变你所创建的世界。只需在浏览世界时输入新的提示词,你就可以实时编辑你的环境。改变天气、添加角色或重定向剧情。Genie 4 首次将这些视觉效果与实时原生音频相结合,随着场景展开,音频从环境声景到角色对话都会随之适应变化。

Discover new Genie 3 capabilities

发现新的 Genie 3 功能

While Genie 4 delivers our highest-fidelity visuals, physics, and native audio, Genie 3 remains available in Project Genie’s model picker when you want longer runs or panorama-seeded spaces. Today’s Genie 3 updates include:

虽然 Genie 4 提供了最高保真度的视觉效果、物理引擎和原生音频,但当你需要更长的运行时间或基于全景图生成的空间时,Genie 3 仍可在 Project Genie 的模型选择器中使用。今天的 Genie 3 更新包括:

  • Extended sessions and world chaining: Explore for up to 3 minutes per generation (up from 60 seconds), or link multiple worlds together for longer journeys.
  • 360-degree panorama image grounding: Seed new worlds from 360-degree panorama images.
  • Real-time in-world events: Dynamically change the environment with natural-language prompts while you explore.
  • 扩展会话与世界链式连接:每次生成探索时间长达 3 分钟(从之前的 60 秒提升),或将多个世界链接在一起以进行更漫长的旅程。
  • 360 度全景图像锚定:基于 360 度全景图像生成新世界。
  • 实时世界内事件:在探索过程中,通过自然语言提示动态改变环境。

Genie 3 uses real-time promptable events to connect different worlds into one continuous flight across forests, underwater ruins, and icy seas.

Genie 3 利用可实时提示的事件,将不同的世界连接成一次连续的飞行之旅,穿越森林、水下遗迹和冰海。

See how 360-degree panorama grounding in Genie 3 keeps Petra’s canyon and facade consistent as you look around.

看看 Genie 3 中的 360 度全景锚定技术如何在环顾四周时保持佩特拉(Petra)峡谷和立面的连贯性。

Watch how Genie 3 combines longer runs, 360-degree panorama grounding, and real-time promptable events - transforming the world as you fly and even summoning a dragon to ride on command.

观看 Genie 3 如何结合更长的运行时间、360 度全景锚定和可实时提示的事件——在你飞行时改变世界,甚至应命令召唤一条龙供你骑乘。

Create new interactive experiences

创建新的互动体验

The new Project Genie agent takes creation beyond a single text or image prompt. Now crafting an interactive universe is as effortless as having a conversation. Simply describe your vision in natural language, and the agent orchestrates Google’s suite of multimodal models to craft initial visuals establishing your world’s signature look, paired with custom background scores tailored to the mood.

新的 Project Genie 智能体将创作过程超越了单一文本或图像提示的范畴。现在,构建一个互动宇宙就像聊天一样轻松自如。只需使用自然语言描述你的愿景,该智能体便会协调 Google 的多模态模型套件,打造确立世界标志性外观的初始视觉内容,并搭配契合氛围的定制背景配乐。

You can also bring Gemini models into the session as you explore: it observes what is happening on screen, surfaces context-aware suggestion chips for where to go next, or triggers in-world events automatically as your journey unfolds.

在探索过程中,你还可以引入 Gemini 模型参与会话:它会观察屏幕上发生的情况,提供上下文感知的建议芯片以指引下一步行动,或者随着旅程展开自动触发世界内的事件。

Turns a natural language prompt into an interactive spacewalk, then an alien spaceship appears on command

将自然语言提示转化为互动太空行走,随后应命令出现一艘外星飞船

Remix a hillside cafe into a glass pavilion in the desert.

将山坡咖啡馆改造成沙漠中的玻璃亭。

Publish custom worlds to the gallery for others to explore and remix.

将自定义世界发布到画廊,供他人探索和重新混合。

Explore the new gallery

探索新画廊

When you just want to jump straight into exploring, the new Project Genie gallery brings together a rich collection of curated worlds alongside creative work from the community. Transform personal photos into explorable spaces in “Living Memory,” wander through rich prehistoric habitats grounded in 360-degree panoramas in “Dino Field Guide,” or pit themed objects against one another on a desk with Gemini serving as your real-time referee in “Rock Paper Scissors.”

当你只想直接开始探索时,新的 Project Genie 画廊汇集了大量精心策划的世界以及来自社区的创意作品。在“Living Memory”中将个人照片转化为可探索的空间;在“Dino Field Guide”中漫步于基于 360 度全景的丰富史前栖息地;或在“Rock Paper Scissors”中让主题物体在桌面上相互对抗,由 Gemini 担任你的实时裁判。

Browse curated and community-built worlds and jump straight into exploring.

浏览精心策划和社区构建的世界,直接开始探索。

“Villa Giallo,” created with Will aka BLVCKL!GHT, uses Genie 4 to turn a gothic manor into an interactive, room-by-room mystery.

“Villa Giallo”由 Will(即 BLVCKL!GHT)使用 Genie 4 创建,将一座哥特式庄园转变为逐室解谜的互动体验。

“The Life of Ants,” built with filmmaker Anand Gandhi, uses Genie 4’s photorealism to let you navigate a colony from an ant’s perspective.

“The Life of Ants”与导演 Anand Gandhi 合作打造,利用 Genie 4 的 photorealism(照片级真实感)让你以蚂蚁的视角探索一个蚁群。

The gallery also showcases inspiring collaborations with pioneering creators who had early access to these tools. Teaming up with creative technologist Will aka BLVCKL!GHT, we created “Villa Giallo,” an atmospheric, room-by-room mystery inside a gothic manor. And with filmmaker Anand Gandhi, we built “The Life of Ants,” inviting you to navigate a complex colony from an ant’s perspective, drawing on Genie 4’s stunning photorealism.

画廊还展示了与早期获得这些工具访问权限的先驱创作者合作的鼓舞人心的案例。我们与创意技术专家 Will(又名 BLVCKL!GHT)联手创作了“Villa Giallo”,这是一部在哥特式庄园内逐室展开的氛围悬疑作品。同时,我们与导演 Anand Gandhi 合作打造了“The Life of Ants”,邀请你以蚂蚁的视角探索复杂的蚁群,并借助 Genie 4 令人惊叹的照片级真实感。

Beyond exploring curated worlds, you can also launch your vision with a simple prompt or remix existing gallery experiences alongside the Project Genie agent — tweaking the underlying world structure, soundtrack, or rules to make it uniquely your own. Once you’ve crafted something special, publish your creation to the community gallery for others to explore and remix.

除了探索精心策划的世界,你还可以使用简单的提示词启动你的愿景,或与 Project Genie 代理一起混音现有的画廊体验——调整底层世界结构、配乐或规则,使其独一无二。一旦你创作出特别的内容,就可以将其发布到社区画廊,供他人探索和混音。

Learn how we are building responsibly

了解我们如何负责任地进行构建

We are committed to developing and rolling out interactive world models responsibly. All video and audio generated in Project Genie includes SynthID watermarking, and all model inputs and outputs go through safety screenings.

我们致力于负责任地开发和部署交互式世界模型。Project Genie 中生成的所有视频和音频均包含 SynthID 水印,且所有模型输入和输出都经过安全筛查。

Genie 3 and Genie 4 remain world models in the frontier research space. As the research continues to evolve, we are actively tackling key challenges such as camera latency, intricate physical interactions, real-time audio precision, and long-term scene permanence. For a deeper look into model limitations and our progress on these fronts, read our Genie 4 model research overview.

Genie 3 和 Genie 4 仍处于前沿研究领域的世界模型阶段。随着研究的不断演进,我们正在积极解决关键挑战,如相机延迟、复杂的物理交互、实时音频精度以及长期场景持久性。如需深入了解模型局限性及我们在这些方面的进展,请阅读我们的 Genie 4 模型研究概述。

Try Project Genie today

立即试用 Project Genie

Eligible Google AI Ultra subscribers can try the new Project Genie experience starting today, and Google AI Pro members can join the waitlist on the same page.

符合条件的 Google AI Ultra 订阅用户可从即日起体验全新的 Project Genie 功能,Google AI Pro 成员可在同一页面加入等待名单。

Get the latest news from Google in your inbox

将 Google 的最新新闻发送至您的收件箱

更进一步:量化金融体系

看懂新闻只是起点——沿量化金融路径,把它变成能交付的工程能力

进入量化体系 →