NVIDIA Unlocks AI Compute at Scale, Inviting Partners to Power the AI Infrastructure Buildout
NVIDIA introduces a revenue-sharing model to expand AI compute access for startups, enterprises, and AI clouds, enabling faster deployment of AI factories with Grace Blackwell GB300 GPUs.
The useful question is what changes for users, developers or buyers, and whether the announcement stays industry context or becomes something people can actually use.
NVIDIA is shifting its business model to address growing AI compute demands by offering a revenue-sharing and credit-support framework for AI clouds. This initiative allows partners to procure NVIDIA infrastructure for AI-native customers, aligning economic incentives with usage. The model aims to accelerate adoption of NVIDIA platforms while providing recurring earnings linked to cloud revenue. AI clouds and enterprises benefit from faster access to full-stack accelerated computing without delays in infrastructure setup.
Sharon AI and Firmus are among the first partners deploying NVIDIA-powered AI factories under this new model. Sharon AI plans to deploy up to 40,000 NVIDIA Grace Blackwell GB300 GPUs, emphasizing sovereign AI compute infrastructure. Firmus is constructing a DSX AI factory campus in Batam, Indonesia, scaling to 360 megawatts and 170,000 GPUs. Both companies highlight the need for scalable, cost-efficient compute to support global AI competition.
The initiative targets AI-native companies such as Baseten, Fireworks AI, and Together AI, which require immediate access to AI cloud capacity for training, fine-tuning, and inference. These companies face challenges in securing reliable, large-scale NVIDIA accelerated computing as their usage grows. The new model provides commercial flexibility to transition from pilot to production deployments. AI clouds and enterprises can now access compute resources without traditional capital-intensive barriers.
NVIDIA’s announcement underscores the broader shift from model development to production-scale AI inference. By enabling AI clouds to sell NVIDIA-powered services, the company aims to unlock financing for compute infrastructure while supporting high-growth AI sectors. The model also introduces a recurring revenue stream for NVIDIA tied to cloud usage, reinforcing its role in the AI infrastructure ecosystem.