Google 第八代 TPU 新增 TPU 8i Max 与 TPU 8tx
Four distinct TPU systems in a single generation in one year! Instead of waiting years for a traditional semiconductor cycle, hardware engineers co-designing sy
Google 扩展第八代 TPU 产品线，新增面向超大上下文窗口与单机架最高算力密度的 TPU 8i Max，以及面向安全机密推理的 TPU 8tx。Amin Vahdat 称这是 Google 首次在一年内、同一代际推出四款不同的 TPU 系统。开发者可通过 TorchTPU 直接运行原生 PyTorch 与 Hugging Face 模型且无需改写代码，开源 TPU Sync 将 KV 传输延迟降低 90% 以上，Agent Compute Engine 用于减少闲置算力浪费。
推荐理由
AI 评分 72
Google 一年内推出四款 TPU 系统，并给出 PyTorch 与 Hugging Face 模型免改写迁移路径，可观察硬件迭代节奏对开发者的影响。
主题
查看原帖非常高兴与大家分享我们 @Google 的 AI 与基础设施团队取得的一项重大里程碑。 今天，我们扩展了第八代 TPU 产品组合，新增： • TPU 8i Max：专为超大上下文窗口和单机架最高算力密度而打造 • TPU 8tx：专为安全、机密推理而设计 这是我们首次在一年内、同一代际推出四款不同的 TPU 系统。 对开发者而言，硬件的价值取决于软件栈。我们希望你们能够像我们在 Google 一样轻松高效地使用 TPU 进行构建。现在你可以： ↳ 通过 TorchTPU 运行原生 PyTorch 与 Hugging Face 模型（无需改写代码） ↳ 借助开源 TPU Sync 将 KV 传输延迟降低 90% 以上 ↳ 通过 Agent Compute Engine 消除闲置算力浪费 在我们的博客上了解更多关于最新 TPU 的信息：https://t.co/8eaFhq3jhi 我期待与不断拓展 AI 与计算边界的开发者、研究人员、工程师和组织建立联系！
查看原文
Thrilled to share a big milestone from our AI & Infrastructure team at @Google. Today we’re expanding our 8th generation TPU portfolio with: • TPU 8i Max: Built for massive context windows and the highest compute density per rack • TPU 8tx: Purpose-built for secure, confidential inference This marks the first time we’ve introduced four distinct TPU systems in a single generation, in one year. For developers, hardware is only as good as the software stack. And we want you to be able to build with TPUs as easily and efficiently as we do at Google. You can now: ↳ Run native PyTorch & Hugging Face models via TorchTPU (zero code rewrites) ↳ Cut KV transfer latency >90% with open-source TPU Sync ↳ Eliminate idle compute waste with Agent Compute Engine Read more about our newest TPUs on our blog: https://t.co/8eaFhq3jhi I look forward to connecting with builders, researchers, engineers and organizations pushing the boundaries of AI and computing!
正文
Four distinct TPU systems in a single generation in one year! Instead of waiting years for a traditional semiconductor cycle, hardware engineers co-designing systems alongside Google Deepmind means we can keep innovating quickly for customers.