Google introduces TPU 8tx and 8i Max and makes Cloud Pathways generally available
Google introduced two TPU systems, the 8tx for confidential inference and the 8i Max, and made Cloud Pathways generally available on Google Kubernetes Engine.

Watch the short
Under a minute, read by an AI anchor. Made .
YouTube loads only when you press play.
Google has introduced two TPU systems. The 8tx is aimed at confidential inference, meaning running models on data that should stay protected during processing. The 8i Max packs two chips into one package for workloads that use large context windows, the span of text a model can consider at once.
Google also made Cloud Pathways generally available on Google Kubernetes Engine. Google describes it as the distributed runtime it uses for Gemini. The available material gives no benchmark results, pricing or availability dates for either TPU system, so the claims rest on their names and intended uses.
What changed
Google introduced the TPU 8tx for confidential inference and the 8i Max, which packs two chips into one package, and made Cloud Pathways generally available on Google Kubernetes Engine.
Type: New capability · Applies to: TPU 8tx, 8i Max, Cloud Pathways, Google Kubernetes Engine
Why it matters
Companies running large-context or confidential inference on Google Cloud gain new hardware options, and Pathways is now available for their own distributed workloads.
What you can do
Evaluate the 8i Max and 8tx against your current accelerators for long-context or confidential inference workloads.
Original source
We have no link to the original source for this story yet.
Further reporting
Where we saw it
Forward Future (Matthew Berman), Turing Post
AI wrote this summary from the sources listed, and nobody reviewed it individually before publication. Check the original source before you rely on it. How it works