Neocloud

A cloud provider that specialises in renting out computing infrastructure for AI, mainly servers fitted with GPUs. Uptime Intelligence describes neoclouds as "a new category of cloud providers" that "focus on offering GPU-backed servers and virtual machines, often at prices more affordable than those of the hyperscalers", naming CoreWeave, Nebius and Lambda Labs among them. In its February 2025 price comparison, an on-demand Nvidia DGX H100 instance averaged $98 an hour from a hyperscaler against $34 from a neocloud. It notes that neoclouds largely sell access to the same third-party chips, so they compete mainly on price, and that hyperscalers are customers too: Microsoft agreed to spend $10 billion on infrastructure through CoreWeave by the end of 2029. For businesses, a neocloud can lower the cost of training and running AI, but it adds another supplier, contract and data-transfer question to a cloud strategy.