In a significant move to address the growing GPU shortage plaguing AI development, NVIDIA announced DGX Cloud Lepton at Computex 2025 in Taiwan on May 19. The platform creates a unified marketplace connecting AI developers with GPU resources from multiple cloud providers worldwide.
The timing is crucial as GPU demand for AI training and deployment has reached unprecedented levels. According to Gartner analyst Chirag Dekate, "Not just NVIDIA, but NVIDIA's supply chain can barely keep up with the demand, and demand far exceeds supply." DGX Cloud Lepton aims to solve this bottleneck by aggregating GPU resources from various providers.
Ten NVIDIA Cloud Partners (NCPs) including CoreWeave, Crusoe, Lambda, and SoftBank will offer NVIDIA Blackwell and other architecture GPUs through the marketplace. Developers can access compute capacity in specific regions for both on-demand and long-term needs, supporting strategic and sovereign AI operational requirements.
"NVIDIA DGX Cloud Lepton connects our network of global GPU cloud providers with AI developers," said Jensen Huang, founder and CEO of NVIDIA. "Together with our NCPs, we're building a planetary-scale AI factory."
The platform integrates with NVIDIA's software stack, including NIM and NeMo microservices, Blueprints, and Cloud Functions, to accelerate AI application development and deployment. It also provides management software for cloud providers with real-time GPU health diagnostics and automated root-cause analysis.
Beyond just connecting resources, DGX Cloud Lepton offers a unified experience across development, training, and inference workflows. Developers can purchase GPU capacity directly from participating providers or bring their own compute clusters, enabling deployment across multi-cloud and hybrid environments with minimal operational burden.