Google Cloud has introduced 2 new custom AI chips as part of its 8th-generation tensor processing units (TPUs), aimed at improving both model training and inference workloads.
The new lineup includes TPU 8t, designed for AI model training, and TPU 8i, built for inference, which refers to real-time usage of trained models after users submit prompts.
According to Google, the new chips deliver up to 3x faster model training, 80% better performance per dollar, and the ability to scale beyond 1 million TPUs in a single cluster. The company says this will enable higher compute capacity with lower energy consumption and reduced costs for customers.
While the development strengthens Google’s in-house AI capabilities, it is not positioned as a direct replacement for Nvidia. Instead, Google continues to integrate Nvidia hardware into its cloud infrastructure and plans to offer Nvidia’s upcoming Vera Rubin chip later this year.
Like other hyperscalers such as Microsoft and Amazon, Google is building its own AI chips to complement, not replace, existing GPU-based systems.
Industry observers note that while cloud providers may gradually reduce dependence on third-party chips, Nvidia remains dominant in the AI hardware market. Analyst Patrick Moorhead highlighted that earlier predictions about TPUs disrupting Nvidia have not materialised, with Nvidia now reaching a nearly $5 trillion market valuation.
Google also confirmed continued collaboration with Nvidia to enhance cloud performance. Both companies are working on improving networking efficiency using Falcon, a software-based networking technology developed by Google and open-sourced in 2023 through the Open Compute Project.
The launch reflects Google’s strategy to expand its AI infrastructure capabilities while maintaining strong partnerships within the broader ecosystem.
Also read: Viksit Workforce for a Viksit Bharat
Do Follow: The Mainstream LinkedIn | The Mainstream Facebook | The Mainstream Youtube | The Mainstream Twitter
About us:
The Mainstream is a premier platform delivering the latest updates and informed perspectives across the technology business and cyber landscape. Built on research-driven, thought leadership and original intellectual property, The Mainstream also curates summits & conferences that convene decision makers to explore how technology reshapes industries and leadership. With a growing presence in India and globally across the Middle East, Africa, ASEAN, the USA, the UK and Australia, The Mainstream carries a vision to bring the latest happenings and insights to 8.2 billion people and to place technology at the centre of conversation for leaders navigating the future.
