Tag
9 articles
Google's latest video reveals how Tensor Processing Units (TPUs) are powering increasingly complex AI workloads with unprecedented efficiency and speed.
Google has unveiled its next-generation TPUs, splitting the design into specialized training and inference chips, signaling a new direction in AI hardware development.
Google Cloud launches two new AI chips to compete with Nvidia, offering improved performance and cost efficiency. The company continues to use Nvidia hardware for now.
Google unveiled its 8th-gen TPUs, a revamped agent platform, and a new AI layer for Workspace at Cloud Next '26, all under the 'Agentic Enterprise' banner.
Google has launched two specialized TPUs designed to power the next generation of agentic AI systems, marking a significant advancement in AI hardware capabilities.
This article explains Google's strategy to challenge Nvidia in AI inference by building a diverse chip supply chain with four partners, spanning multiple chip generations from Ironwood to TPU v8.
This article explains the concept of AI chips, how they work, and why companies like Anthropic are exploring custom chip design as a strategic move to optimize performance and control their AI infrastructure.
Anthropic has signed a multi-gigawatt TPU deal with Google and Broadcom to secure computing capacity for AI development, with resources expected to come online in 2027.
Meta has signed a multi-billion dollar deal to rent Google's TPUs, challenging Nvidia's dominance in the AI chip market. The move signals increased competition and innovation in AI hardware.
