Nvidia GPUs are everywhere. Here are the ways companies are accessing them
One industry researcher identified over 300 neoclouds that can provide GPU capacity, up about 55% in less than a year.
The article explores the growing demand for Nvidia GPUs in AI and the expanding ecosystem of companies offering GPU access, including hyperscalers, neoclouds, and alternative solutions.
What happened
The article details how companies are accessing Nvidia GPUs through diverse channels, including major cloud providers, neoclouds, and direct purchases, driven by surging demand for AI workloads. It highlights challenges like supply shortages and the rise of alternative solutions.
Why it matters
Understanding GPU access options is critical for AI developers and businesses to manage costs, scalability, and performance. The article underscores the evolving infrastructure needs of AI workloads and the competitive landscape for GPU providers.
- GPU access is no longer limited to major cloud providers, with neoclouds and alternative models gaining traction.
- Supply constraints and pricing volatility require businesses to diversify GPU sourcing strategies.
- On-premises and hybrid solutions remain viable for organizations prioritizing control and cost management.
- Nvidia's dominance in AI hardware drives growth across the entire GPU ecosystem.
- Tactical partnerships and direct GPU purchases are becoming more common for large-scale AI projects.
The bigger picture
The proliferation of GPU access methods reflects the industry's need to balance cost, scalability, and reliability. Companies must navigate these options to optimize AI development, while GPU providers face pressure to expand capacity and innovation.
The article highlights the expanding ecosystem of GPU access options for AI, driven by surging demand and supply chain challenges, emphasizing the need for flexible and strategic infrastructure planning.
Technical details
- Nvidia B200 GPU spot prices have more than doubled since March 2024.
- Neoclouds like CoreWeave and Nebius offer specialized GPU access with flexible configurations.
- Hyperscalers provide trusted infrastructure but may lack sufficient GPU capacity for large-scale AI projects.
- Bring-your-own-hardware models, as offered by Oracle, allow companies to leverage existing GPU investments.
FAQ
How many neoclouds now provide Nvidia GPU capacity?
Over 300 neoclouds, a 55% increase in less than a year.
Why are companies turning to neoclouds for GPU access?
Neoclouds offer flexibility, specialized configurations, and capacity that hyperscalers may lack, especially for large-scale AI projects.
What role do hyperscalers play in GPU access?
Hyperscalers like Amazon, Microsoft, and Google remain dominant but face supply constraints, prompting companies to seek alternative providers.
How are companies addressing GPU supply shortages?
Through tactical deals, on-premises solutions, neoclouds, and bring-your-own-hardware models to ensure access and manage costs.
What factors influence a company's choice of GPU access method?
Factors include cost, scalability, data center location, technical requirements, and the need for flexibility in GPU availability.
This summary and analysis were generated with AI from the original article and may contain errors. Read the original for the full story.
Read the full story
Originally published by CNBC — Technology on Oct 10, 2026.
Comments
Sign in to join the discussion.