aifollow.news
← TrendingLatest development Nov 10 02:24
STORY IN FOCUS

Google unveils two new TPUs for confidential AI workloads and larger inference tasks

Google has unveiled two eighth-generation TPUs. The TPU 8tx, designed with AMD, is intended to isolate sensitive data and model weights during training and inference. The TPU 8i Max combines the capabilities of two TPU 8i chips in one package, doubling memory capacity, processing power and bandwidth so models can handle more information in queries.

1 developments2 reports2 sources

Event developments

Coverage of the same occurrence shares a node; subsequent developments have their own nodes.

Google expands its eighth-generation TPU portfolio with 8i Max and 8tx

Google announced an expansion of its eighth-generation TPU portfolio: 8i Max targets large context windows and high compute density per rack, while 8tx is designed for confidential inference. Developers can now run native PyTorch and Hugging Face models through TorchTPU; the announcement is aimed at teams building and running AI systems.

View source reports · 2 sources · 2 reports
All reports · 2

By original publication time, with each report's bookmarks and feedback preserved.

TechCrunch AI Machine translation

Google unveils two new TPUs for confidential AI workloads and larger inference tasks

Google has unveiled two eighth-generation TPUs. The TPU 8tx, designed with AMD, is intended to isolate sensitive data and model weights during training and inference. The TPU 8i Max combines the capabilities of two TPU 8i chips in one package, doubling memory capacity, processing power and bandwidth so models can handle more information in queries.

Rohan Paul (@rohanpaul_ai) Machine translation

Google expands its eighth-generation TPU portfolio with 8i Max and 8tx

Google announced an expansion of its eighth-generation TPU portfolio: 8i Max targets large context windows and high compute density per rack, while 8tx is designed for confidential inference. Developers can now run native PyTorch and Hugging Face models through TorchTPU; the announcement is aimed at teams building and running AI systems.