NVIDIA revenue-sharing model opens GPU access for AI clouds

NVIDIA introduced a new business model on July 2, 2026 that lets AI clouds procure its infrastructure through a revenue-sharing and credit-support arrangement. Under the model, AI clouds sell NVIDIA-powered services, and NVIDIA earns standard product revenue plus a share of the cloud revenue on the supported capacity. The structure aims to accelerate adoption among AI-native companies and give NVIDIA a recurring, usage-linked earnings stream.

First partners and capacity

Sharon AI and Firmus are among the first companies working with NVIDIA under this model. Sharon AI is deploying up to 40,000 NVIDIA Grace Blackwell GB300 GPUs. Firmus is building a DSX AI factory campus in Batam, Indonesia, expected to scale to 360 megawatts and up to 170,000 NVIDIA GPUs. James Manning, cofounder and CEO of Sharon AI, said the collaboration supports the company's mission to deliver sovereign, large-scale AI compute infrastructure. Tim Rosenfield, co-CEO of Firmus Technologies, said the DSX-aligned AI factory will help more customers access the compute they need.

What it means for inference providers

The initiative targets the shift from model development to production inference, where compute demand is moving toward continuously operating AI factories that generate tokens at scale. NVIDIA said the model addresses the limited access emerging AI companies historically had to capital-intensive infrastructure, where even long-term commitments were insufficient to unlock financing. For model builders, inference providers, agent platforms and enterprises scaling AI, the model can mean faster access to full-stack accelerated computing without waiting through site selection, power procurement, construction and hardware bring-up.

AI-native companies such as Baseten, Fireworks AI and Together AI show where compute demand is headed, according to NVIDIA. They need immediate access to AI cloud capacity for model training, post-training, fine-tuning and high-volume agentic inference. Their customers need reliable access to large-scale NVIDIA accelerated computing as usage grows, but also commercial flexibility as products move from pilot to production. NVIDIA said the model enables AI clouds to procure infrastructure for AI-native, enterprise and ISV customers through economic alignment.