精选归档 · 第 4 页

6180 条 · 共 312

7月23日7月23日周四

星期四 · 3 条
19:11
Google AI:DEV 作者专属(RSS)精选
AI 评分 61/100
Google Cloud Agent Skills 完整指南:从基础到高级云运维

Google Cloud 推出官方 Agent Skills 开源指令集,旨在让 AI 编码智能体安全、高效地执行多步骤云操作。该仓库基于 agentskills.io 开放标准,目前收录 70 多项技能,涵盖安全审计、无服务器部署、BigQuery 优化等 8 个类别。技能采用渐进式披露模型,通过验证工作流、安全门控和上下文感知机制,防止智能体盲目执行破坏性命令。


推荐理由:这是 Google Cloud Skills 系列教程第一篇,从安装到触发讲得很细,把 Agent Skills 的“渐进式披露”和安全门机制都拆解了,适合想让 AI 编码代理帮你管云资源的开发者。
17:25
Google AI:DEV 作者专属(RSS)精选
AI 评分 73/100
Gemini 3.6 Flash 与 3.5 Flash-Lite 正式版发布

Google 发布 Gemini 3.6 Flash 和 Gemini 3.5 Flash-Lite 正式版。3.6 Flash 在复杂智能体和多模态任务上性能更强,输出 token 价格降至 $7.50/1M,支持 1M token 上下文窗口和 Computer Use 工具。


推荐理由:Gemini 3.6 Flash 降本增效,3.5 Flash-Lite 主打低成本高速,弃用 temperature 等参数是个硬变更,使用 Gemini 的开发者必须检查迁移。

7月22日7月22日周三

星期三 · 5 条
08:00
Tomer Tunguz 博客(VC 分析)精选
AI 评分 75/100
Google Cloud 营收增速向 NVIDIA 看齐

Google Cloud 在 2026 年 Q2 营收同比增长 82% 至 $24.8b,超过 $22.3b 的预期,增速已与 NVIDIA 趋同。增长主要来自 AI 算力租赁需求,而非 TPU 系统销售——后者本季度才开始确认收入,大部分收入将在 2027 年落地。云业务积压订单达 $514b,同比增长 385%,营业利润率从一年前的 20.7% 扩大至 35.6%。


推荐理由:作者用财报数据对比 Google Cloud 与 NVIDIA 的增速趋同和利润率变化,帮助读者理解云厂商与芯片商共享同一需求曲线。
00:00
Google Developers Blog(RSS)精选
AI 评分 63/100
Google 推出 Tunix:基于 JAX 的高吞吐智能体后训练库

Google 发布 Tunix,一个基于 JAX 的原生后训练库,旨在消除多轮、使用工具的 LLM 推理智能体训练中的 TPU 闲置瓶颈。Tunix 通过高并发异步 rollout 与解耦的生产者-消费者流水线最大化硬件吞吐量,确保训练器持续获得数据。该库提供即插即用抽象和持续宏观级性能分析,便于集成自定义环境。


推荐理由:Google 这个训练库直接解决 agent 训练中 TPU 空转的隐性成本,对做后训练的团队可能是基础设施级改进,值得跟进。
00:00
Google Cloud:Databases(RSS)精选
AI 评分 71/100
AlloyDB 推出列式引擎加速 HNSW,pgvector 向量搜索 QPS 提升最高 4.9 倍

Google Cloud 的 AlloyDB 在预览版中推出列式引擎加速的 HNSW 索引,使 pgvector 向量搜索的每秒查询数(QPS)相比标准 PostgreSQL HNSW 提升最高 4.9 倍,在固定 QPS 下召回率可提升 0.163。该加速通过将索引固定在列式引擎内存中、使用向量化内存布局并绕过标准 PostgreSQL 缓冲管理器开销实现,无需修改应用即可获得性能提升。


推荐理由:AlloyDB 把 pgvector 的 HNSW 搜索提速 4 倍,不是靠 GPU 而是靠内存架构,对已经在 PostgreSQL 上跑 RAG 的团队可能是个省钱又提精度的实用升级,点开看看基准测试。

7月21日7月21日周二

星期二 · 1 条
23:22
Google DeepMind:Blog(RSS)精选
AI 评分 57/100
Google DeepMind 发布 Gemini 3.6 Flash、3.5 Flash-Lite 与 3.5 Flash Cyber 三款新模型

Google DeepMind 推出 Gemini 3.6 Flash、3.5 Flash-Lite 和 3.5 Flash Cyber 三款新模型。其中 Gemini 3.6 Flash 为最新主力模型,3.5 Flash-Lite 主打更低成本与更高效率,3.5 Flash Cyber 则针对网络安全场景优化。三款模型均通过 Google AI 开发者平台提供 API 访问。


推荐理由:Gemini Flash 系列常规更新,Flash-Lite 和 Cyber 变体值得 API 开发者留意,但细节还没出来,目前只能先标记。

7月17日7月17日周五

星期五 · 1 条
00:32
Google Blog:AI(RSS)精选
AI 评分 60/100
Google Vids 上线 Gemini Omni 与个人数字分身功能

Google Vids 推出两项更新:Gemini Omni 支持用户通过自然语言提示词和图片参考生成、逐步编辑高质量视频片段;个人数字分身功能允许用户上传自拍和语音录制后,输入文字即可让数字分身出镜。两项功能面向 Google AI Pro 和 Ultra 订阅者及 Google Workspace 商业客户开放,所有生成内容均含不可见的 SynthID 数字水印。


推荐理由:如果你已经在用 Google Workspace,这两个更新让 Vids 的可用性提升了一档,但对于外部的 AI 视频生成,没带来新变量。

7月16日7月16日周四

星期四 · 4 条
22:27
Google AI:DEV 作者专属(RSS)精选
AI 评分 72/100
DiffusionGemma 开发者指南发布

Google AI 发布 DiffusionGemma 开发者指南,该实验性模型基于 Gemma 4 架构,采用计算受限并行生成,在单张 NVIDIA H100 上实现 1000+ tokens/秒的生成速度。模型为 26B 参数的 MoE 架构,推理时仅激活 3.8B 参数,可在 18 GB VRAM 内量化部署。


推荐理由:虽然 DiffusionGemma 是实验性的,但把文本生成从内存带宽瓶颈转向计算瓶颈的思路非常新颖,这篇指南给出了可落地的服务方案和微调手法,做推理优化的值得跟进。
20:10
The Verge:AI(RSS)精选
AI 评分 78/100
欧盟裁定 Google 必须向竞争对手开放 Android 和 Search,影响 Gemini 等 AI 服务

欧盟依据《数字市场法案》(DMA)裁定 Google 必须向竞争对手开放 Android 和 Google Search 的关键部分,包括允许第三方 AI 助手和搜索引擎获得更大访问权限。这两项决定可能削弱 Google 对两大核心平台的控制,并为其 AI 工具 Gemini 的未来格局带来深远影响,同时为竞争对手创造新的发展机会。


推荐理由:欧盟这两项裁决直接撬开了谷歌安卓和搜索的围墙,对Gemini等AI助手的地位是实质冲击,做AI应用和搜索的创业者值得盯着后续执行细节。
17:58
Google Developers Blog(RSS)精选
AI 评分 60/100
Gemini Enterprise Agent Platform 新增 Parallel Web Search 网络接地提供商

Google Cloud 与 Parallel Web Systems 合作,将 Parallel 的搜索基础设施作为新的网络接地提供商原生集成到 Gemini Enterprise Agent Platform 中。该集成使开发者能将 AI 智能体锚定在可验证的实时网络结果上,以提升企业工作流的事实准确性。用户还可编程提取、永久缓存并与其他大语言模型一起处理网络数据。


推荐理由:为 Gemini Agent 平台引入第三方搜索 grounding,企业开发者多了一个直接可用的实时事实核查选项,虽非重大突破,但对做合规和知识库场景的团队是个实用更新。
04:02
Hacker News 热门(buzzing.cc 中文翻译)精选
AI 评分 73/100
前谷歌DeepMind研究员因公司签署无限制军事AI协议而离职

前谷歌DeepMind研究员Alex Turner因谷歌向国土安全部出售云服务并最终签署无限制军事AI协议而离职。他曾起草25页提案要求加入禁止杀手机器人和大规模监控的合同条款,但提案被CEO转交后无人跟进。Turner指出,包括Jeff Dean和Stuart Russell在内的多位AI伦理领袖在关键时刻未能兑现承诺。


推荐理由:Alex Turner用亲身经历戳穿了AI巨头们的伦理承诺,Jeff Dean、Stuart Russell等名人在关键时刻失声,这份记录比任何声明都真实。

7月15日7月15日周三

星期三 · 2 条
02:35
TechCrunch:AI(RSS)精选
AI 评分 70/100
Google 因 AI 训练再遭出版商集体诉讼

包括 Hachette、Cengage、Elsevier 及作家 Scott Turow 在内的出版商与作者团体对 Google 提起集体诉讼,指控其未经授权使用受版权保护的作品训练 Gemini 模型,并故意移除或篡改版权信息以掩盖这一行为。原告称 Google 将原本仅用于 Google Books 搜索片段展示的书籍副本,以及 Google Play 商店上传的图书,非法用于 AI 训练。诉讼援引 Google 内部文件,其中指出此举可能带来“100 亿至 1000 亿美元的潜在罚款”。该案在纽约南区联邦地区法院提起。


推荐理由:出版商集体诉讼谷歌,这次带上了内部文件警告‘可能面临千亿美元罚款’的弹药,是版权方对AI训练最有力的一次反击,值得关注后续发展。
00:51
Google Developers Blog(RSS)精选
AI 评分 58/100
Google 在 I/O Connect India 展示由 Tensor SoC 和 TPU 驱动的 Pixel 10 端侧 AI 未来

在 Google I/O Connect India 上,Google 展示了由定制 Tensor SoC 和 TPU 驱动的 Pixel 10 系列所支持的 100% 私有端侧 AI 未来。活动首次推出轻量级 Gemma 4 E2B 模型,该模型原生运行于设备端,可实现完全离线的多模态功能,包括 AI 聊天、实时图像识别和个人智能体任务。开发者即日起可通过新发布的 Tensor SDK beta 及其配套开源资源,开始构建这些安全的边缘应用。


推荐理由:我觉得这是端侧 AI 从实验到落地的信号,Gemma 4 E2B 的离线能力搭配新 SDK,让完全本地的多模态应用不再是期货,做移动隐私 AI 的开发者该上手试试了。

7月14日7月14日周二

星期二 · 4 条
23:33
Google AI Developers@googleaidevs精选
AI 评分 70/100
Google AI 发布 Gemini 3.5 Live Translate,支持 70+ 语言近实时语音到语音翻译http://x.com/i/article/2067347903761485824How Developers are Building with Gemini 3.5 Live Translate70+ languages. Fluid, natural speech-to-speech. Near real-time latency.Gemini 3.5 Live Translate is fundamentally changing how developers build global, multilingual applications. By processing raw audio streams rather than converting text turn-by-turn, it preserves speaker intonation, pacing, and pitch.Southeast Asia's leading superapp, Grab, is exploring how to break down language barriers between drivers and travelers. With users making over 10 million voice calls a month, the model makes cross-lingual communication flow naturally.You can build and deploy your own voice translation apps by using the Gemini Live API with LiveKit, Fishjam, Pipecat, or Vision Agents. These integrations handle the complex real-time media streaming infrastructure, so you can focus on the user experience.Take a closer look at how you build with these tools ⬇️LiveKit (@livekit)By combining LiveKit Agents with Gemini 3.5 Live Translate, the LiveKit team built a virtual meeting room where everyone speaks their native language and instantly understands each other, breaking down language barriers. The model handles the multilingual inputs automatically and streams speech continuously, delivering fluid translations just a few seconds behind the speaker.Software Mansion (@swmansion)The Software Mansion team paired Gemini 3.5 Live Translate with the ultra-fast MoQ (Media over QUIC) protocol. This integration bypasses traditional streaming bottlenecks, setting a new standard for delivering high-quality, speech-to-speech translation to large audiences with exceptionally low latency.VisionAgents AI (@visionagents_ai)Real-time translation is challenging, but switching between different languages dynamically is even harder. Watch as VisionAgents AI tests the limits of the new model by throwing multiple languages at it on the fly. Using Gemini 3.5 Live Translate's auto-detection, the agent effortlessly handles the rapid context switching while maintaining natural intonation, rather than falling back on robotic, turn-based responses.Ready to build?You can try Gemini 3.5 Live Translate in Google AI Studio and grab starter code from the Gemini Cookbook. Read the blog for more details.Google AI 发布 Gemini 3.5 Live Translate,支持 70+ 语言、近实时延迟的语音到语音翻译。该模型直接处理原始音频流,保留说话者语调、节奏和音高。东南亚超级应用 Grab 正探索将其用于司机与乘客间的跨语言沟通,其用户每月发起超 1000 万次语音通话。开发者可通过 Gemini Live API 集成 LiveKit、Fishjam、Pipecat 或 Vision Agents 构建应用。LiveKit 已实现虚拟会议室多语言即时理解;Software Mansion 结合 MoQ 协议突破流媒体瓶颈;VisionAgents AI 展示了动态多语言切换能力。开发者可在 Google AI Studio 试用并获取 Cookbook 示例代码。
推荐理由:Google把Gemini 3.5的语音翻译做到了近实时、保持语调,这对出海产品是个很实际的升级,实时通话翻译终于从demo走向可用。
23:06
Google AI:DEV 作者专属(RSS)精选
AI 评分 68/100
DiffusionGemma 发布:文本扩散模型实现 4 倍加速推理

Google 发布实验性开源模型 DiffusionGemma,采用文本扩散技术并行生成 256 token 块,推理速度最高提升 4 倍。该 26B MoE 模型仅激活 3.8B 参数,量化后适配 18GB VRAM 消费级 GPU,在单张 H100 上达 1000+ tokens/s。


推荐理由:Google把扩散模型首次做进大语言模型,生文速度提升4倍,还开源了权重。虽然实验性质量打折,但本地实时交互的开发者终于有个新选项可以玩了。
17:57
Google AI:DEV 作者专属(RSS)精选
AI 评分 66/100
Google ADK 2.4.0 发布:Agent 可直接触发动态工作流

Google 发布 ADK 2.4.0,允许将 Workflow 直接注册为 Agent 的工具列表中的一等工具,使协调 Agent 能自动调用动态工作流。该模式通过 @node 装饰器定义动态执行节点,支持在运行时根据用户输入的任务列表迭代调用子 Agent 生成执行计划。示例使用 gemini-3.5-flash 模型实现任务协调,代码已开源至 devrel-demos 仓库。


推荐理由:ADK 2.4 把动态工作流直接注册成 Agent 的工具,补齐了多智能体编排里缺失的那块拼图,做复杂 Agent 的可以直接抄这个模式,代码拿来就能跑。
17:32
Demis Hassabis@demishassabis精选
AI 评分 68/100
Demis Hassabis:AGI 数年可至,影响达工业革命10倍http://x.com/i/article/2076946210397552640A Framework for Frontier AI and the Dawning of a New AgeThis is a pivotal moment in human history. Artificial General Intelligence (AGI), a system that exhibits all the cognitive capabilities the brain has, is probably only a few short years away. When we look back on this time in the decades to come, I think we will realise we were standing in the foothills of the singularity - nothing less than the dawning of a new age for humanity.I’ve spent my whole life working on AGI because I’ve always had a deep conviction that, if built and deployed responsibly, it would prove to be one of the most beneficial and transformative technologies ever invented. AGI cannot be compared to standard technological breakthroughs, not even ones as consequential as the internet or mobile - it is much more akin to the discovery of electricity or fire. If you stop to think about it, we’ve essentially found a way to make sand think. It’s miraculous.The magnitude of this technology’s impact will be unprecedented, perhaps 10x of the Industrial Revolution at 10x the speed. It will help us solve some of the biggest problems society faces from accelerating drug discovery to developing new clean energy sources to creating novel advanced materials. We could even reach a point where resources are no longer the limiting factor for human progress, leading to an amazing new era of abundance.The Challenges of the FrontierAI is already starting to deliver real-world benefits but to realise its immense promise, we have to navigate this critical period of development thoughtfully and carefully. Urgent action is needed to address risks that might arise as we get closer to AGI. We’ve already seen the challenges frontier models pose for cybersecurity, and other threats including nuclear and bio risks may soon emerge as capabilities continue to advance. On the horizon, we will need robust safeguards to maintain control of increasingly agentic, recursively self-improving systems - and tackle unknown issues that will only become clearer over time.I’ve always believed in the power of human ingenuity and creativity to solve any problem. I’m confident that mitigating the technical risks related to AI is a challenge we can collectively address, but only if we give ourselves the time and space to get this next crucial step right. Currently, as a field and as a wider society, we aren’t doing that.At the moment, we are locked in an extremely intense, multilayered commercial and geopolitical race. While these competitive dynamics fuel rapid progress and accelerate the incredible upsides, advances on the frontier are outpacing our understanding of the technology. Nobody in the world knows for sure what is going to happen from here, and even the experts disagree. When there is a large degree of uncertainty and the stakes are this high, proceeding with cautious optimism is the sensible and correct strategy. That calls for public policy that promotes innovation while also incentivising responsibility and security, fosters international collaboration on key safety issues, and encourages careful consideration of how AI is deployed for the benefit of society.A Framework for a Frontier AI Standards BodyThe rapid progress we’re seeing in AI requires a new approach to testing frontier AI model capabilities that is dynamic, adaptable, and rigorous. The US is well positioned, given its economic and technical standing, to take the first step in developing such a framework. It could establish a new Standards Body modelled on a federally overseen public-private partnership or self-regulatory organisation, much like the Financial Industry Regulatory Authority (FINRA), with a board that includes independent leading technical experts and open-source representatives. Funding would need to be substantial and likely mostly come from industry, in order to attract world-class technical talent and provide the necessary compute resources for large-scale testing.The Standards Body would be responsible for developing assessment protocols and working with appropriate federal agencies and the US National Labs to conduct testing in areas relevant to national security. A model would qualify as ‘Frontier-class’ if it meets certain thresholds on a set of benchmarks determined by the Standards Body and regularly updated to keep pace with evolving AI capabilities. Organisations with ‘Frontier Models’ as defined by those benchmarks would be deemed ‘Frontier Labs’, and be encouraged to adopt best practices, such as publishing model cards with technical details, maintaining strong internal cybersecurity, vetting key personnel, and providing sufficient resourcing for safety and security research, and more.Initially, Frontier Labs would voluntarily share models with the Standards Body for review up to 30 days before release. Once the assessment protocol is shown to be effective and robust, formalisation could quickly follow, meaning that Frontier Models would be required to pass it to be deployed in the US market. Labs would also work with the Standards Body to address any critical post-release vulnerabilities.Model assessments should include rigorous scientific evaluations of capabilities in cybersecurity, biological threats and other high-risk domains. Specific agentic AI tests could look for attempts to bypass safety guardrails or signs of deception, and ensure best practices, such as digitally watermarking AI-generated images and generating human-readable output tokens to understand model reasoning.These evaluations would be regularly updated, perhaps quarterly to start, with outdated or saturated benchmarks being deprecated and replaced. Initially, they would be developed in consultation with Frontier Labs, but eventually the Standards Body should build up the technical capacity to create its own held-out tests independent of the Labs to prevent overfitting. Working with the US government, it could promote an ecosystem of third-party auditors to help with the assessments and development of new benchmarks and evaluations.The strength of this approach is it would be technically focused, while at the same time supporting innovation and incentivising responsible behaviour. It is designed to keep up with the field’s acceleration and adapt to the biggest risks as they are identified, and could be ratcheted up if the seriousness of the situation demands, including coordinating a slowdown in development among the Frontier Labs if deemed necessary. Being designated a Frontier Lab would carry significant prestige and be open to any organisation by building models that meet the benchmark criteria. The framework could apply to Frontier-class models no matter their country of origin or whether they are open or closed, but any non-frontier models, say from startups or academia, would be exempt from this process.This US-initiated effort would provide a strong starting point for creating shared international standards on Frontier AI. Since this technology is going to affect the entire planet, ideally this framework would spur the international community to reach a consensus on how to manage the most serious risks while ensuring everyone has access to and can benefit from the opportunities that AI brings.The Future Is Not Yet WrittenAGI has the potential to be the ultimate tool for advancing science and medicine, and to drive enormous productivity gains and economic growth. But in order to achieve this, we need to get the technical foundations right by coordinating around a shared global framework, using the most rigorous scientific methods, and bringing the best minds together to work on the challenges we face.Even if we solve these hard technical challenges, there will be further complex economic and philosophical questions to tackle: what sorts of new economic models will be needed to help everyone thrive in a post-scarcity world? What values do we want to live by, what will meaning and purpose be, and how might even the human condition itself change? Resolving these questions obviously cannot and should not be left to technologists alone. It requires every part of society to come together to help define this new chapter.There is both huge excitement and uncertainty around AI, and both are warranted. But the future is not yet written, we must use this precious window before AGI arrives to shape this technology for the benefit of all humanity. What we collectively do now will determine how the next phase of civilisation unfolds. By safely stewarding AGI into the world, we can enter a new golden age of scientific discovery and progress, and usher in a bright future of incredible human flourishing.Google DeepMind 联合创始人 Demis Hassabis 发文称,AGI 可能仅需数年即可实现,其影响将达工业革命的10倍且速度更快。他指出,前沿模型在网络安全、核与生物风险方面已构成挑战,未来需对日益智能体化、递归自我改进的系统建立稳健防护。Hassabis 呼吁美国率先建立类似 FINRA 的前沿AI标准机构,采用联邦监督下的公私合作或自律组织模式,由独立技术专家和开源代表组成董事会,资金主要来自行业以吸引顶尖人才和算力。他强调,当前商业与地缘竞赛导致技术进步快于理解,需以谨慎乐观态度推进公共政策,兼顾创新与安全。
推荐理由:Demis Hassabis 亲自下场提出一个具体的 AGI 监管框架,用 FINRA 模式构建标准组织,这比泛泛呼吁更有行动感,政策讨论里少见的可操作方案。