Nvidia to Create Marketplace for Cloud-Based AI chips
Nvidia announces a new software platform to create a marketplace for cloud-based artificial intelligence chips. Nvidia's graphics processors, or GPUs, dominate the market for training AI models, and a raft of new cloud players called neoclouds, such as CoreWeave and Nebius Group, have emerged to specialize in renting out Nvidia's chips to software developers.
The company, based in Santa Clara, California, revealed a new tool named Lepton that enables cloud computing firms to offer GPU capacity in a single location.
Alongside CoreWeave and Nebius, other companies participating in the Lepton platform include Crusoe, Firmus, Foxconn, GMI Cloud, Lambda, Nscale, SoftBank Corp, and Yotta Data Services.
Nvidia cloud Vice President Alexis Bjorlin states that although there is a significant increase in demand for chips from both startups and large corporations, locating available chips has been a highly manual process.
Alexis states, "It feels as if everyone is contacting everyone regarding the available computing resources. We aim to ensure a seamless experience, as it promotes the growth and development of the ecosystem, allowing all clouds—both global and emerging cloud providers—to tap into Nvidia's complete developer ecosystem."
So far, Nvidia's Lepton partner list does not include major cloud providers like Microsoft, Amazon Web Services, or Alphabet's Google. Bjorlin mentioned that the system is created for them to sell their capacity in the marketplace if they wish.
Bjorlin says that Lepton will ultimately enable developers to locate Nvidia chips in particular countries to fulfill data storage needs. It will enable companies that already possess certain Nvidia chips to more easily look for additional ones to rent.
Also Read: WAVES 2025: Burgeoning India's Orange Economy
Nvidia did not reveal what the business model for the new software platform will be, nor did it specify if it would impose commissions or fees on developers or cloud services.