Hardware
We're launching two specialized TPUs for the agentic era.
Google announced Ironwood, its seventh-generation TPU that is twice as power efficient as the previous generation, alongside specialized hardware designed to support the emerging agentic AI era. Ironw
Google announced Ironwood, its seventh-generation TPU that is twice as power efficient as the previous generation, alongside specialized hardware designed to support the emerging agentic AI era. Ironwood is purpose-built for demanding workloads including large-scale model training, reinforcement learning, and high-volume low-latency AI inference, operating four times faster than Google's sixth-generation TPU Trillium. Google also introduced updates to its Axion CPU-powered N4A instances to enable tight orchestration between general-purpose CPUs and AI accelerators, offering twice the price-performance ratio of comparable x86-based VMs.
Related
- Two new TPUs to power the next wave of AI training and inference at Google
- Sources: Google is in talks with Marvell Technology to develop a memory processing unit that works alongside TPUs, and a new TPU for running AI models (Qianer Liu/The Information)
- SK hynix says it has begun mass production of the 192GB SOCAMM2, a next-gen LPDDR5X low-power DRAM module designed particularly for Nvidia's Vera Rubin (The Korea Herald)
- Generative Design for Direct-to-Chip Liquid Cooling for Data Centers
Source: Google AI | 2026-04-22