Google 发布 Gemini 3.7 Flash、Gemini 3.5 Transcribe 和 Pixel 11 系列等八月 AI 更新

在过去的20多年里，我们在机器学习和人工智能研究、工具及基础设施方面持续投资，旨在打造能够改善更多人日常生活的产品。谷歌各团队正在探索在医疗、危机响应和教育等广泛领域释放人工智能效益的方法。为了让你了解我们的最新进展，我们将定期汇总谷歌最近的人工智能新闻。

以下是我们对八月份部分人工智能公告的回顾。

在八月份，我们继续负责任地推动人工智能的发展——使其更快速、更易获取，并真正为每个人提供实用价值，无论是软件开发人员、学生，还是创意工作者和科学家。我们将智能直接带到人们的工作和生活场景中：推出专为Gemini设计的Pixel 11系列强大硬件，推出如Gemini 3.7 Flash这样的高性价比开发者模型，在Android上推出Chrome中的Gemini：https://blog.google/products-and-platforms/products/chrome/gemini-in-chrome-android-auto-browse/，以及覆盖Google Workspace和Gemini Live的免提语音工具。随着Gemini应用正式突破每月10亿用户，我们还扩大了创意和科学领域的影响力，推出了工作室级视频和音乐生成，以及预测天气模式和应对气候挑战的开源人工智能模型。总的来说，八月份标志着人工智能从理论上的强大向日常现实中的实用性转变。

使用Gemini 3.7 Flash以更低成本构建更强大的智能代理：https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/。我们发布了Gemini 3.7 Flash，这是迄今为止最智能的工作型模型，专为编码和智能代理打造。它在3.6 Flash发布仅三周后推出，在软件工程、知识工作及网页开发工作流程方面实现了显著提升——每百万令牌的入门价格仅为原3.6 Flash成本的一半。

试试全新的 Pixel 11 系列，专为 Gemini 智能设计：https://blog.google/products-and-platforms/devices/pixel/google-pixel-11-pro-xl/。在 Made by Google 2026 活动上，我们发布了 Pixel 11、Pixel 11 Pro、Pixel 11 Pro XL 和 Pixel 11 Pro Fold。这些设备配备了重大相机升级、增强的耐用性以及我们速度最快、功能最强的芯片 Google Tensor G6，可运行最新的 Gemini Nano 模型。1 它们还专为 Gemini 智能设计，提供节省时间的个人帮助。2 查看 Made by Google 2026 的所有公告：https://blog.google/products-and-platforms/devices/pixel/made-by-google-2026/。

用我们赠送的一年 Gemini 开始学期：https://blog.google/innovation-and-ai/products/gemini-app/student-offer-google-ai/。我们为全球符合条件的大学生提供一年免费的 Google AI 计划 —— 还包括新的和增强的学习工具 —— 帮助学生充分利用这个学年。我们还提供帮助教师和学生开学的工具：https://blog.google/products-and-platforms/products/education/back-to-school-2026/，其中包括专门的学生中心、新的教师主导工具：https://blog.google/products-and-platforms/products/education/iste-2026-educator-updates/，以及 Gemini 中的 SAT 备考：https://blog.google/innovation-and-ai/products/gemini-app/how-to-take-practice-sat/。

用搜索提升你的学习：https://blog.google/products-and-platforms/products/search/back-to-school-study-tools/。为了帮助你自信地开始学期，我们在搜索中添加了新的 AI 驱动学习功能 —— 所有功能均从设计上保证安全。通过这些更新，搜索可以通过互动视觉帮助你理解复杂概念，为 SAT、ACT、GRE 和 LSAT 等考试生成练习测验，使用 Lens 进行逐步学习，并通过笔记本保持学习有条理。

使用 Gemini 3.5 转录获得更智能的转录服务：https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5-transcribe/。我们最新的语音转文本模型为开发者工作流（如语音助手、实时字幕和通话后分析）提供精准、智能的实时转录。不同于在噪音和行话环境中表现不佳的传统模型，Gemini 3.5 转录能将原始音频直接转换为准确、精炼、兼具上下文理解的内容。

通过 Gemini Live 的新生产力功能提高工作效率：https://blog.google/innovation-and-ai/products/gemini-app/productivity-features-gemini-live/。Gemini Live 正在超越对话，帮助您处理更复杂的任务。借助个人智能、每日简报、Spark 和免手操作收件箱管理等新功能，您可以轻松讨论一天的安排并委派待办事项，不会错过任何重要内容。

看看超过10亿人如何每月使用 Gemini 应用：https://blog.google/innovation-and-ai/products/gemini-app/one-billion-monthly-users/。Gemini 应用正式突破每月10亿用户，成为谷歌历史上增长最快的产品。为纪念这一里程碑，我们分享了一些使用洞察，例如：63%的用户现在直接与 Gemini 对话，包括更多“纯语音”用户——繁忙父母使用它处理日常事务的可能性提高了43%。Gemini 目前每天生成超过1.5亿张图片，小型企业是其主要用户，依赖 Gemini 的一站式图片、视频和音频创作功能制作营销材料。

使用 Gemini Omni 1.1 Flash 生成更可控的视频：https://blog.google/innovation-and-ai/technology/developers-tools/build-with-gemini-omni-1-1-flash/。我们推出了 Gemini Omni 1.1 Flash，为人们提供生成视频时更高的精确度和控制力。新功能可实现工作室级的视频制作——包括场景扩展、首尾帧插值、清晰的 4K 放大以及更快的原型制作。Omni 1.1 Flash 现已在 Google Flow：https://blog.google/innovation-and-ai/models-and-research/google-labs/new-creative-controls-google-flow/、Google AI Studio：https://aistudio.google.com/prompts/new_chat?model=gemini-omni-1.1-flash、Gemini 企业代理平台：http://console.cloud.google.com/agent-platform/overview，以及 Gemini 应用：https://gemini.google.com/ 上可用。

探索 Gemma 如何在从太空到水下的各个环境中离线使用：https://blog.google/innovation-and-ai/technology/developers-tools/gemma-one-billion-downloads/。我们发布了 Gemma，帮助开发者在任何地方构建负责任且创新的 AI 应用。一亿次下载之后，Gemma 支持从手机和边缘基础设施到太空等各种环境。你可以探索人们使用它的方式——从研究跨物种交流到推动医学突破——并在我们的新社区资源库中分享、发现和协作。

了解“蓝天行动”，这是一个利用 AI 减少航空气候影响的项目：https://blog.google/innovation-and-ai/models-and-research/google-research/blue-skies/。我们的 AI 驱动预报已经帮助飞行机组人员和空中交通管制员调整航线，以避免形成飞机尾迹——所有操作都在正常飞行范围内。现在，我们正与英国政府及航空领导者合作，将这项技术扩展到北大西洋，帮助航空公司在全球范围内减轻航空对气候的影响。

看看 WeatherNext 2 在预测气旋方面如何实现了巨大的进步：https://blog.google/innovation-and-ai/models-and-research/google-deepmind/weathernext-2-cyclones/。在《自然》杂志的一篇论文中，我们的研究人员展示了 WeatherNext 2 如何以最先进的准确性预测气旋的路径、强度和风结构——在一个模型中实现了十年的气象进展。现在，我们将 WeatherNext 2 开源给研究社区，以帮助建立全球气候韧性。

检查你的收件箱以确认你的订阅。

与谷歌 Tensor G5 发布时相比。

仅对部分国家和语言的 18 岁以上用户可用。功能可用性会有所不同；某些功能可能需要订阅以获得更高使用量。请检查响应。部分功能可通过 Gemini 应用提供。

For more than 20 years, we’ve invested in machine learning and AI research, tools, and infrastructure to build products that make everyday life better for more people. Teams across Google are working on ways to unlock AI’s benefits in fields as wide-ranging as healthcare, crisis response, and education. To keep you posted on our progress, we're doing a regular roundup of Google's most recent AI news.

Here’s a look back at some of our AI announcements from August.

In August, we continued advancing AI responsibly — making it faster, more accessible, and truly practical for everyone, from software developers and students to creatives and scientists. We’re bringing intelligence directly to where people work and live, with powerful new hardware designed for Gemini in the Pixel 11 series, cost-efficient developer models like Gemini 3.7 Flash, the rollout of Gemini in Chrome：https://blog.google/products-and-platforms/products/chrome/gemini-in-chrome-android-auto-browse/ on Android, and hands-free voice tools across Google Workspace and Gemini Live. As the Gemini app officially crossed 1 billion monthly users, we also expanded our creative and scientific footprint, introducing studio-quality video and music generation alongside open-source AI models that predict weather patterns and tackle climate challenges. Overall, August marked a shift toward AI that isn't just powerful in theory, but useful in everyday reality.

Build better agents at a lower cost with Gemini 3.7 Flash ：https://blog.google/innovation-and-ai/models-and-research/gemini-models/introducing-gemini-3-7-flash/. We released Gemini 3.7 Flash as our most intelligent workhorse model yet for coding and agents. It arrived just three weeks after our launch of 3.6 Flash, delivering substantial improvements across software engineering, knowledge work, and web development workflows — with an introductory price of half the original 3.6 Flash cost per million tokens.

Try the new Pixel 11 series, designed for Gemini Intelligence ：https://blog.google/products-and-platforms/devices/pixel/google-pixel-11-pro-xl/. At Made by Google 2026, we unveiled Pixel 11, Pixel 11 Pro, Pixel 11 Pro XL, and Pixel 11 Pro Fold. These devices come with major camera upgrades, enhanced durability, and our fastest, most powerful chip, Google Tensor G6, that runs the latest Gemini Nano model. 1 They’re also designed for Gemini Intelligence to deliver time-saving, personal help. 2 See all the announcements from Made by Google 2026：https://blog.google/products-and-platforms/devices/pixel/made-by-google-2026/.

Start the semester with one year of Gemini, on us ：https://blog.google/innovation-and-ai/products/gemini-app/student-offer-google-ai/. We’re offering one year of a Google AI plan free of charge for eligible college students around the world — plus new and enhanced study tools — so students can make the most of this school year. We’ve also got tools to help teachers and students as they head back to school：https://blog.google/products-and-platforms/products/education/back-to-school-2026/, including a dedicated student hub, new teacher-led tools：https://blog.google/products-and-platforms/products/education/iste-2026-educator-updates/, and SAT prep in Gemini：https://blog.google/innovation-and-ai/products/gemini-app/how-to-take-practice-sat/.

Level up your learning with Search ：https://blog.google/products-and-platforms/products/search/back-to-school-study-tools/. To help you start the semester with confidence, we've added new AI-powered learning features in Search — all built to be safe by design. With these updates, Search can help you grasp complex concepts through interactive visuals, generate practice quizzes for exams like the SAT, ACT, GRE, and LSAT, learn step-by-step with Lens, and stay organized with notebooks.

Get more intelligent transcription with Gemini 3.5 Transcribe ：https://blog.google/innovation-and-ai/models-and-research/gemini-models/gemini-3-5-transcribe/. Our latest speech-to-text model delivers precise, intelligent real-time transcription for developer workflows like voice agents, live captioning, and post-call analytics. Unlike conventional models that struggle with noise and jargon, Gemini 3.5 Transcribe converts raw audio directly into accurate, polished, context-aware understanding.

Get more done with new productivity features in Gemini Live ：https://blog.google/innovation-and-ai/products/gemini-app/productivity-features-gemini-live/. Gemini Live is moving beyond conversation to handle complex tasks on your behalf. With new features like Personal Intelligence, Daily Brief, Spark, and hands-free inbox management, you can easily talk through your day and delegate your to-dos without missing a beat.

See how more than 1 billion people are using the Gemini app every month ：https://blog.google/innovation-and-ai/products/gemini-app/one-billion-monthly-users/. The Gemini app officially surpassed 1 billion monthly users, making it the fastest-growing product in Google’s history. To mark the milestone, we shared some usage insights, such as: 63% percent of users now talk directly to Gemini, including more “voice only” users — with busy parents 43% more likely to use it for everyday tasks. Gemini now generates 150 million+ images every day, and small businesses are power users, relying on Gemini's all-in-one image, video, and audio creation to craft marketing materials.

Generate videos with more control using Gemini Omni 1.1 Flash ：https://blog.google/innovation-and-ai/technology/developers-tools/build-with-gemini-omni-1-1-flash/. We introduced Gemini Omni 1.1 Flash to bring people even more precision and control for generating videos. The new capabilities deliver studio-quality video production — including scene extension, first-and-last-frame interpolation, crisp 4K upscaling, and faster prototyping. Omni 1.1 Flash is now available in Google Flow：https://blog.google/innovation-and-ai/models-and-research/google-labs/new-creative-controls-google-flow/, Google AI Studio：https://aistudio.google.com/prompts/new_chat?model=gemini-omni-1.1-flash, the Gemini Enterprise Agent Platform：http://console.cloud.google.com/agent-platform/overview, and the Gemini app：https://gemini.google.com/.

Explore how Gemma is offline everywhere from outer space to underwater ：https://blog.google/innovation-and-ai/technology/developers-tools/gemma-one-billion-downloads/. We released Gemma to help developers build responsible, innovative AI applications anywhere. Over one billion downloads later, Gemma supports environments from phones and edge infrastructure to space. You can explore some of the ways people are using it — from researching interspecies communication to driving medical breakthroughs — and share, discover, and collaborate in our new community repository.

Learn about Operation Blue Skies, a project to reduce aviation climate impact with AI ：https://blog.google/innovation-and-ai/models-and-research/google-research/blue-skies/. Our AI-powered forecasts already help flight crews and air traffic controllers adjust routes to avoid forming contrails — all within normal flight operations. Now, we’re partnering with the UK Government and aviation leaders to expand this technology across the North Atlantic, helping airlines reduce aviation’s climate impact on a global scale.

See how WeatherNext 2 demonstrated a massive leap forward in predicting cyclones ：https://blog.google/innovation-and-ai/models-and-research/google-deepmind/weathernext-2-cyclones/. In a Nature paper, our researchers showed that WeatherNext 2 predicts cyclone track, intensity, and wind structure with state-of-the-art accuracy — delivering a decade of meteorological progress in one model. Now, we’re open-sourcing WeatherNext 2 to the research community to help build global climate resilience.

Check your inbox to confirm your subscription.

Compared to Google Tensor G5 at launch.

Available for select countries and languages to users 18+. Feature availability will vary; some features may require subscription for higher usage. Check responses. Some features available through Gemini app.