Introducing DGX Cloud Lepton and the AI Compute Marketplace
NVIDIA has announced
NVIDIA DGX Cloud Lepton, an innovative AI platform designed to democratize access to high-performance computing resources through a global marketplace. This platform is set to connect developers building both agentic and physical AI applications to vast pools of GPUs supplied by a worldwide network of cloud providers.
A Global Network of Cloud Partners
To deliver the necessary computing capacity required for rapid AI advancement, NVIDIA is collaborating with a diverse range of Cloud Partners—known as NCPs—including CoreWeave, Crusoe, Firmus, Foxconn, GMI Cloud, Lambda, Nebius, Nscale, Softbank Corp., and Yotta Data Services. These partners will make their latest GPU architectures, including the cutting-edge
Blackwell series, available through the
DGX Cloud Lepton marketplace. This alliance enables developers to:
- Access GPU compute resources on demand or for long-term projects
- Select computing power in specific regions to meet strategic and sovereign AI operational needs
- Leverage offerings from leading cloud service providers and GPU marketplaces
Unified Access for AI Developers
DGX Cloud Lepton tackles the persistent industry challenge of securing reliable and high-performance GPU resources. By unifying access to cloud AI services and GPU capacity, the platform provides developers with comprehensive infrastructure to accelerate and scale their AI innovations. Integration with the broader NVIDIA software stack, including AI development and optimization tools, ensures robust support and seamless workflow.
Benchmarking and Optimization Tools
The platform offers benchmarking tools like
NVIDIA DGX Cloud Benchmarking—a suite that helps optimize AI workload performance and measure cost-to-performance ratios. Such capabilities make it easier for startups and enterprises to quantify ROI and plan their AI strategies effectively.
Geographic Expansion: Asia-Pacific Highlights
Yotta Data Services has emerged as the first key NVIDIA Cloud Partner in the Asia-Pacific region to participate in the
NVIDIA Exemplar Cloud initiative. This step represents a significant stride towards providing greater geographic diversity of compute capacity.
Enrollment and Further Information
Early access for developers is now available; interested parties can
sign up for NVIDIA DGX Cloud Lepton. The official unveiling took place during the COMPUTEX keynote, part of a series of global events including
NVIDIA GTC Taipei.
Market Impact
By creating a marketplace for AI compute power, NVIDIA is tackling the scaling bottleneck facing AI developers and enterprises. The DGX Cloud Lepton model, with unified, region-specific access to GPU resources, is positioned to serve the increasing demand for strategic AI deployment and innovation worldwide.
