Nvidia Expands AI Access with Global GPU Cloud Marketplace
| 2025-05-19
Nvidia has launched a new cloud service, DGX Cloud Lepton, aimed at giving artificial intelligence developers worldwide broader access to its powerful GPUs through a network of cloud providers.
Nvidia launches DGX Cloud Lepton, a GPU marketplace connecting AI developers with global cloud providers, expanding access beyond major platforms.
Nvidia has launched a new cloud service, DGX Cloud Lepton, aimed at giving artificial intelligence developers worldwide broader access to its powerful GPUs through a network of cloud providers.
The announcement was made during the Computex technology conference in Taiwan on Monday. Nvidia said the platform allows developers to select from a range of cloud vendors to train and deploy AI models, easing dependence on traditional hyperscale cloud service providers such as Amazon Web Services, Microsoft Azure, or Google Cloud.
Marketplace model for AI infrastructure
DGX Cloud Lepton functions as a marketplace, offering developers the flexibility to choose from Nvidia’s network of global GPU cloud providers. These include firms such as CoreWeave, Lambda, and Crusoe. Nvidia said this model simplifies the often complex process of accessing high-performance computing resources for artificial intelligence applications.
“Nvidia DGX Cloud Lepton connects our network of global GPU cloud providers with AI developers,” said Jensen Huang, chief executive of Nvidia.
By aggregating excess GPU capacity from various cloud firms, Nvidia aims to create a more efficient and scalable ecosystem for AI workloads. The company believes this marketplace can lower barriers for enterprises and researchers that may otherwise face difficulty securing sufficient computing resources.
Addressing demand surges in AI computing
The demand for graphics processing units (GPUs), critical for training large-scale AI models, has soared since late 2022 amid rapid advancements in generative AI and machine learning technologies. Nvidia’s GPUs have become especially sought after due to their performance and compatibility with modern AI workloads.
"Many developers, whether in research or enterprise, struggle to find available compute," said Alexis Bjorlin, vice president of Nvidia’s DGX Cloud division. “Lepton reduces that friction by making it easier to identify and access available GPU capacity.”
This marketplace approach could help smaller cloud vendors utilise underused resources. According to Nvidia, providers like CoreWeave can list available GPU capacity, enabling developers to find computing power without relying exclusively on larger platforms.
Direct outreach to AI developers
The launch also marks a strategic shift for Nvidia, which is increasingly moving to engage developers and enterprises directly rather than solely through cloud service intermediaries. Analysts say the move could help Nvidia strengthen its position in the AI ecosystem and diversify its customer base.
“This is Nvidia’s way to kind of be an aggregator of GPUs across clouds,” said Ben Bajarin, chief executive of market-research firm Creative Strategies.
Developers using DGX Cloud Lepton can choose to work with a single provider or across multiple vendors depending on their specific needs, Nvidia said. “It is up to the developer to choose,” Bjorlin noted. “Nvidia is not intersecting in that pathway.”
Context: Nvidia’s rising role in cloud and AI
Nvidia, once known primarily for gaming graphics cards, has become a cornerstone of the artificial intelligence revolution. The company's high-performance GPUs are used to train and run models behind technologies such as ChatGPT, autonomous driving systems, and medical diagnostics.
Its entry into cloud services—traditionally dominated by firms like Amazon, Google, and Microsoft—signals an ambition to play a more direct role in shaping the AI infrastructure landscape. By offering developers flexible access to computing power via DGX Cloud Lepton, Nvidia is positioning itself as a central enabler of AI development worldwide.
Industry observers say this approach could reduce developers’ reliance on any single cloud provider and foster more competitive pricing in the market for AI infrastructure. With cloud computing costs a major concern for startups and research institutions, the availability of a flexible GPU marketplace may help accelerate innovation.
DGX Cloud Lepton is currently being rolled out through Nvidia’s global partners, with additional cloud service providers expected to join the network in the coming months.
US President Donald Trump imposes 30% tariffs on imports from the European Union and Mexico from 1 August
US President Donald Trump announces 30% tariffs on EU and Mexico from 1 August, threatening higher taxes if retaliated. The EU and Mexico prepare countermeasures.
| 2025-07-13
Federal Judge Blocks Trump's Birthright Citizenship Order Nationwide
A US judge halts President Trump’s order ending birthright citizenship, granting nationwide protection for children born to non-citizens.
| 2025-07-11
Starlink Approved to Launch Satellite Internet Services in India
India has authorised Starlink to operate satellite broadband, marking a milestone in expanding internet connectivity with Elon Musk’s low Earth orbit network.
| 2025-07-10
Trump Warns Brazil of 50% Tariff Over Bolsonaro Trial and Digital Policies
Donald Trump has threatened Brazil with a 50% tariff from August, citing the trial of Jair Bolsonaro and alleged attacks on US tech firms.
| 2025-07-10
Trump Warns BRICS Nations Including India of 10% Tariff Over Dollar Challenge
Donald Trump has announced a 10% tariff on BRICS countries, claiming the bloc threatens the US dollar's status as the global standard currency.
| 2025-07-09