Google has expanded its eighth-generation TPU portfolio with TPU 8tx and TPU 8i Max, marking four new TPU systems in a single year. The new chips are built for the agentic AI era, supporting massive context windows and secure, confidential inference for enterprise workloads.
At Google Cloud Next in April, Google introduced TPU 8t for large-scale training and TPU 8i for real-time reasoning. TPU 8t offers up to 2.7 times better training performance, while TPU 8i delivers 80% better inference price performance versus the previous generation. TPUs now serve nearly 100 customers, including frontier labs, trading firms, retailers, and universities.
TPU 8i Max targets massive context windows and the highest compute density per rack. TPU 8tx is designed for secure and confidential inference. The expanded eighth-generation portfolio includes TPU 8t, TPU 8i, TPU 8tx, and TPU 8i Max. Google says it is increasingly incorporating feedback from external organizations as computing moves from human to agent scale.
By combining modular custom silicon with an open, easy-to-use software stack, Google aims to deliver specialized supercomputers in months rather than years, making TPUs more accessible to every developer and Cloud customer.
Source: Hacker News · Summarized by HeadlinesBriefing