How we used Gemini to build Google I/O 2026
How we used Gemini to build Google I/O 2026
我们是如何利用 Gemini 构建 Google I/O 2026 的
Jun 01, 2026 2026 年 6 月 1 日
From the jellyfish pre-show to our “TPU Training Day” film, see how Gemini helped make I/O happen this year. 从水母开场秀到我们的“TPU 训练日”短片，一起来看看 Gemini 是如何助力今年 I/O 大会的。
Google I/O 2026 was all about how we’re making AI helpful for everyone in new ways. But we didn’t just make announcements about our innovations in AI at I/O — we used those tools to bring I/O to life, too. Google I/O 2026 的核心主题在于我们如何以全新的方式让 AI 造福每一个人。但我们不仅是在 I/O 大会上发布 AI 创新成果，我们还利用这些工具让 I/O 大会本身变得更加生动。
It’s both a strange and exciting moment to be building anything. We’re living through an incredible shift where AI tools are getting better each month, effectively rewriting the rules of what we can create. 现在是一个既陌生又令人兴奋的创造时代。我们正经历着一场不可思议的变革，AI 工具每个月都在进步，有效地改写了我们创作的规则。
This year, we challenged ourselves to use the same AI we were putting on stage to out-innovate, out-create and out-efficient ourselves. 今年，我们挑战自我，利用我们在舞台上展示的同款 AI 技术，力求在创新、创作和效率上实现自我超越。
We moved faster than ever and prototyped in real-time — blending human artistry with experimental technology — with no better example than the “Timmy TPU” film. 我们以前所未有的速度行动，并进行实时原型设计——将人类艺术与实验性技术相结合——“Timmy TPU”短片就是最好的例证。
But the reward is showing how these tools unlock creativity and offload the mundane tasks, giving the team their best hours back for the parts they are uniquely suited to do. When done right, the event is amazing on its own, and, as a viewer, you stop thinking about how AI was used. That shift is the opportunity we want to share, because people keep asking, “What can you really do with AI?” 但最棒的回报在于，这些工具释放了创造力并分担了繁琐的任务，让团队能将最宝贵的时间投入到他们最擅长的工作中。如果运用得当，活动本身就会非常精彩，作为观众，你甚至不会去思考 AI 是如何被使用的。这种转变正是我们想要分享的机遇，因为人们一直在问：“你到底能用 AI 做什么？”
Keep reading to learn which AI tools we used — and how we prompted them — to help make I/O 2026 happen. 请继续阅读，了解我们使用了哪些 AI 工具，以及我们是如何通过提示词（Prompt）来助力 I/O 2026 的。
AI x Film: “TPU Training Day” short film
AI 与电影：“TPU 训练日”短片
The AI products & models: Google AI Studio; experimental DeepMind models; Gemini Omni; Nano Banana 所用 AI 产品与模型： Google AI Studio；DeepMind 实验性模型；Gemini Omni；Nano Banana
What we did: We created a short film starring a bunch of TPUs getting ready to do some heavy lifting for I/O 2026. 我们做了什么： 我们制作了一部短片，讲述了一群 TPU 为 I/O 2026 准备承担重任的故事。
How we did it: This project started with a question: Could we make an animated film with the simplest materials — cardboard and markers — and then use AI to bring it to life? We worked with director Laurie Rowan and Nexus Studios to blend puppets, traditional animation and AI — keeping human craft and artistry right at the heart of “TPU Training Day” (also known as “Timmy TPU”). 我们是如何做到的： 这个项目始于一个问题：我们能否用最简单的材料（纸板和记号笔）制作一部动画片，然后利用 AI 让它“活”起来？我们与导演 Laurie Rowan 及 Nexus Studios 合作，将木偶、传统动画和 AI 融合在一起，确保人类的工艺和艺术性始终是“TPU 训练日”（又称“Timmy TPU”）的核心。
First, we captured character performances through puppetry and simple 3D animation. This gave us full control over framing and camera movement. We then used Nano Banana to generate stylized first frames from that raw footage. To keep frames consistent, we built a custom tool inside Google AI Studio. This let us test Nano Banana frames at scale, ensuring pixel-perfect matches before generating sequences. We merged the base animation and stylized frames using Gemini Omni and other experimental models. This elevated the film to a cinematic level while retaining the original human intent. Preserving these tiny, human imperfections is what gives puppet films their charm, and our AI pipelines were designed to protect those details. 首先，我们通过木偶戏和简单的 3D 动画捕捉角色表演。这使我们能够完全控制构图和摄像机运动。然后，我们使用 Nano Banana 从原始素材中生成风格化的首帧。为了保持帧的一致性，我们在 Google AI Studio 中构建了一个自定义工具。这使我们能够大规模测试 Nano Banana 的帧，确保在生成序列之前实现像素级的匹配。我们使用 Gemini Omni 和其他实验性模型将基础动画与风格化帧进行了合并。这在保留人类原始创作意图的同时，将影片提升到了电影级水准。保留这些细微的人类瑕疵正是木偶电影的魅力所在，而我们的 AI 工作流正是为了保护这些细节而设计的。
AI x Visual Design: I/O visual brand identity
AI 与视觉设计：I/O 视觉品牌标识
The AI products & models: Gemini models and Nano Banana 所用 AI 产品与模型： Gemini 模型和 Nano Banana
What we did: We created the visual brand identity for I/O 2026, landing on a four-color gradient with overlapping transparencies and interlocking icons. 我们做了什么： 我们为 I/O 2026 创建了视觉品牌标识，最终确定了采用带有重叠透明度和互锁图标的四色渐变设计。
How we did it: Our brand identity was a close collaboration between our team and AI. We started by feeding Gemini models our past brand guidelines and five years of I/O recaps. The early outputs didn’t quite hit the mark, so we ran some micro-experiments. We generated new imagery and iteratively fed outputs back into Nano Banana with feedback. We also used Nano Banana to explore icon styles. Finally, we landed on flat 2D icons that dynamically transform into hyper-textured 3D icons. This created a cohesive brand expression across keynotes, physical signage and digital apps. 我们是如何做到的： 我们的品牌标识是团队与 AI 紧密合作的结晶。我们首先将过去的品牌指南和五年的 I/O 回顾喂给 Gemini 模型。早期的输出结果并不理想，因此我们进行了一些微实验。我们生成了新的图像，并将输出结果反复反馈给 Nano Banana 进行迭代。我们还使用 Nano Banana 探索了图标风格。最终，我们确定了能够动态转换为超纹理 3D 图标的扁平化 2D 图标。这在主题演讲、实体标牌和数字应用中创造了统一的品牌表达。
AI x Immersive Experiences: The I/O pre-show
AI 与沉浸式体验：I/O 开场秀
The I/O pre-show: Jellectronica I/O 开场秀：Jellectronica
The AI products & models: Google Antigravity; Google Colab; Google CoralNPU; Google Flow Music; Lyria 3 Pro 所用 AI 产品与模型： Google Antigravity；Google Colab；Google CoralNPU；Google Flow Music；Lyria 3 Pro
What we did: We kicked off the pre-show with Jellectronica, a generative music experiment in partnership with the Monterey Bay Aquarium that translated moon jelly movements into sound with Lyria 3 Pro. 我们做了什么： 我们以“Jellectronica”开启了开场秀，这是一个与蒙特利湾水族馆合作的生成式音乐实验，通过 Lyria 3 Pro 将月亮水母的游动转化为声音。
How we did it: We trained a YOLO8 model in Google Colab, and then ran it on Google’s Coral NPU. This tracked jellyfish movement to control the music, which was made using Google Flow Music and the Lyria API. For example, more jellies in the bass section meant louder, more energetic bass. We also vibe-coded a mass stem generator in Google Antigravity to automate the production of music stems, like bass, chords, melody and drums. 我们是如何做到的： 我们在 Google Colab 中训练了一个 YOLO8 模型，然后在 Google 的 Coral NPU 上运行。它通过追踪水母的运动来控制音乐，而音乐则是使用 Google Flow Music 和 Lyria API 制作的。例如，低音区的水母越多，低音就越响亮、越有活力。我们还在 Google Antigravity 中编写了一个大规模音轨生成器，以自动化生产低音、和弦、旋律和鼓点等音乐分轨。
The I/O pre-show: Infinite Scaler and Code the Countdown I/O 开场秀：Infinite Scaler 和 Code the Countdown
The AI products & models: Google AI Studio; Gemini API; Gemini Canvas; Google Antigravity; Lyria 3; Nano Banana 所用 AI 产品与模型： Google AI Studio；Gemini API；Gemini Canvas；Google Antigravity；Lyria 3；Nano Banana
What we did: Infinite Scaler, another part of the pre-show, was a video game where players competed and generated the levels as they played. 我们做了什么： 作为开场秀的另一部分，Infinite Scaler 是一款电子游戏，玩家可以在游玩过程中进行竞争并生成关卡。
How we did it: We wanted players to build infinite 3D worlds quickly just using 2D image generation. To do this, we used Nano Banana to generate sprite sheets from user prompts and reference images via the Gemini API. We sent foreground elements back to Nano Banana to generate normal, roughness and emission maps. This inferred depth, letting us map textures to a 3D cardboard box rendered in WebGL before adding them to the global stack of worlds. We used Google AI Studio for rapid prototyping before moving to Google Antigravity for development; we generated in-game music entirely using Lyria 3. 我们是如何做到的： 我们希望玩家仅通过 2D 图像生成就能快速构建无限的 3D 世界。为此，我们使用 Nano Banana 通过 Gemini API 根据用户提示词和参考图像生成精灵图（sprite sheets）。我们将前景元素发回 Nano Banana 以生成法线贴图、粗糙度贴图和自发光贴图。这推断出了深度信息，使我们能够将纹理映射到 WebGL 渲染的 3D 纸板盒上，然后再将其添加到全球世界堆栈中。我们使用 Google AI Studio 进行快速原型设计，随后转入 Google Antigravity 进行开发；游戏内音乐完全使用 Lyria 3 生成。