NVIDIA Unveils New Business Model, Invites Capital Partners to Build AI Compute at Scale

As AI transitions from model development to production inference, compute demand is accelerating and shifting toward continuously operating AI factories that generate tokens at scale. This shift requires access to large-scale, multi-tenant accelerated computing that can come online quickly, stay highly utilized, and support the economics of token-scale AI services. Emerging AI companies have historically had limited access to capital-intensive infrastructure, with even long-term commitments insufficient to unlock financing for compute.

To address this, NVIDIA introduced a new business model this week that opens compute access to the fast-growing AI ecosystem of startups, model builders, enterprises, research organizations, and regional AI players. Under the model, NVIDIA partners with AI cloud companies to deploy large-scale accelerated computing clusters powered by NVIDIA DSX AI factories that manufacture tokens at scale.

AI cloud companies sell cloud services delivered through these factories while aligning economics through a revenue-sharing and credit-support model with NVIDIA. Sharon AI and Firmus are among the first companies to work with NVIDIA on this initiative. Sharon AI is deploying up to 40,000 NVIDIA Grace Blackwell GB300 GPUs, while Firmus is building a DSX AI factory campus in Batam, Indonesia, expected to scale to 360 megawatts and up to 170,000 NVIDIA GPUs.

For model builders, inference providers, agent platforms, and enterprises scaling AI, the model means 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: they need immediate access to AI cloud capacity for training, fine-tuning, and high-volume agentic inference.

The initiative effectively transforms AI infrastructure from a capital-intensive purchasing model to a usage-based service model, lowering upfront barriers for AI companies while creating a recurring revenue stream for NVIDIA. As AI inference workloads continue to grow, this revenue-sharing compute model could become a significant industry trend.

Watch for whether more AI cloud providers join the program, whether the model effectively alleviates the AI compute shortage, and how this partnership approach affects NVIDIA's long-term financial structure.