#TPU

All articles tagged with "TPU"

Here’s how our TPUs power increasingly demanding AI workloads.

Google Pushes TPU Infrastructure as AI Demand Raises the Stakes for Specialized Compute

Google is spotlighting its Tensor Processing Units as demand for larger AI workloads accelerates, framing custom chips as a core advantage in the race to train and run modern models at scale.

Key Takeaways

Google says its TPUs were built specifically for AI workloads more than a decade ago.
The newest TPU generation is described as delivering 121 exaflops of compute power.

DT Editorial Team·Apr 26, 2026·via blog.google

Google launches two eighth-generation TPU designs aimed at the ‘agentic’ AI era

Google says its new TPU 8i and TPU 8t chips are specialized for distinct AI workloads, with one focused on fast agent performance and the other on large-scale model training.

Key Takeaways

Google introduced TPU 8i for fast AI agent workloads and TPU 8t for training large complex models.
The company says the chips are part of a full-stack infrastructure strategy spanning networking, data centers and efficiency.

DT Editorial Team·Apr 22, 2026·via blog.google