Google Unveils TPU 8i Max and TPU 8tx: A Dual-Chip TPU for Massive Context Windows and a Confidential Inference TPU

Google has added two new systems to its eighth-generation TPU lineup. TPU 8i Max targets massive context windows and rack density, while TPU 8tx targets secure, confidential inference. TPU 8i Max is a dual-chip package built from two TPU 8i chips. Both systems arrive alongside the TPU 8t and TPU 8i, which Google announced in April 2026.

TL;DR

• Size: TPU 8i Max packages two TPU 8i chips in a single package. TPU 8tx combines TPU 8t silicon with 5th Gen AMD EPYC host processors.
• Runs on: Google Cloud TPU systems.
• Performance: TPU 8i Max doubles TPU 8i compute, SRAM, and HBM bandwidth. TPU 8tx cuts latency by up to 2x versus TPU 8t.
• Best for: Fitting twice the silicon into the existing rack footprint (TPU 8i Max).
• Limitations: Google has not published absolute specs, benchmarks, prices, or availability regions for either system.

What Did Google Announce?

Google announced both systems today. These are new system configurations rather than new chips: Google reuses its existing eighth-generation silicon and changes the packaging, hosts, memory, and networking around it.

How Does TPU 8i Max Work?

TPU 8i Max attacks the memory wall. Google combines two low-latency TPU 8i chips into a single package, doubling compute, SRAM, and HBM bandwidth compared with a standalone TPU 8i.

Each TPU 8i chip is rated at...

via MarkTechPost