Google ships four distinct TPU systems in one generation in a single year, adding TPU 8i Max and 8tx
sundarpichai · x · 2026-11-10
Google announced two new systems expanding its 8th-gen TPU lineup: TPU 8i Max, built for massive context windows with the highest compute density per rack, and TPU 8tx, purpose-built for secure confidential inference — the first time Google has released four distinct TPU systems in a single generation within one year.
CEO Sundar Pichai credited co-design between hardware engineers and Google DeepMind for breaking free of the traditional multi-year semiconductor cycle. On the software side, developers can now run native PyTorch and Hugging Face models via TorchTPU with zero code rewrites, and the open-source TPU Synthesis work cuts KV transfer latency by over 90%.
Related event: Google Ships Four TPU Systems in One Year, Adding 8tx and 8i Max(10 posts)→
More from Infra
- Cybercriminals Hack AI Inference Provider Fireworks, Customer Data and Models Affected — annatonger · 2026-11-10
- AI's bottleneck shifts from chips to power: model firms spend billions on turbines — TansuYegen · 2026-11-10
- How CrossnoKaye grounds physical AI agents in closed-loop industrial control with Bedrock AgentCore — AWS ML Blog · 2026-11-10
- Gary Marcus slams Nvidia after report says Commerce inquiry was dropped — GaryMarcus · 2026-11-10
- Stratoscope: monitor local LLM token speed and cache hits live in your menu bar — LisaMarieJosefine · 2026-11-10
- Claude Fable 5.2 and Mythos 5.2 spotted on AWS Bedrock, priced 20% below 5.1 — testingcatalog · 2026-11-10