谷歌推出TPU 8tx和TPU 8i Max晶片，用於AI計算
谷歌推出兩款新ai晶片tpu 8tx和tpu 8i max，旨在實現機密推理和大規模ai工作負載，降低對外部晶片製造商的依賴。
google has introduced two new ai chips, tpu 8tx and tpu 8i max, which are part of its eighth generation of tensor processing units tpUs. tpu 8tx, developed with amd, is designed for confidential training and running of ai models, featuring security directly in the hardware layer. tpu 8i max combines two inference-focused tpu 8i chips, doubling memory capacity, processing power, and bandwidth. these chips help google reduce reliance on external chipmakers like nvidia for its cloud computing business. this launch comes as hyperscalers and ai labs are working on their own ai-specific semiconductors.