Top 10: Neocloud Companies Transforming Global Data Centres

The AI boom has forged a new class of cloud providers. Known as neoclouds, these specialised companies are purpose-built from the ground up to deliver the raw, scalable computing power that today’s most demanding AI and high-performance computing (HPC) workloads require.

Unlike the sprawling, general-purpose offerings of hyperscalers like Amazon Web Services or Microsoft Azure, neoclouds focus on one thing: delivering Graphics Processing Unit-as-a-Service (GPUaaS) with unparalleled performance, speed and cost-efficiency. They provide access to top-tier GPUs, often before they are widely available elsewhere, without the queues or complexities of traditional cloud platforms.

For data centre operators and enterprise IT leaders, this emerging ecosystem represents a paradigm shift. 

Neoclouds are not just digital infrastructure suppliers – they are critical partners enabling the AI revolution, from research labs training foundational models to media companies performing real-time rendering. 

Here, we rank the top 10 neocloud players shaping this vital new market.

10. Paperspace

• HQ: New York, USA (now part of DigitalOcean)
• CEO: Dillon Erb (prior to acquisition by DigitalOcean)
• Founded: 2014

Acquired by DigitalOcean in 2023 for $111 million, Paperspace has established itself as a leading platform for building and scaling AI applications.

It provides a complete MLOps software suite that abstracts away the complexities of cloud infrastructure, allowing data scientists to focus on model development rather than server management. 

By integrating Paperspace’s powerful GPU tooling, DigitalOcean has expanded its portfolio to offer a simplified, developer-first cloud that now caters directly to the burgeoning AI and machine learning market.

9. Nscale

• HQ: London, UK
• CEO: Josh Payne (Co-Founder)
• Founded: 2023

Nscale positions itself as a sustainability-led AI cloud, vertically integrating GPU infrastructure with abundant Nordic hydro power. 

The company has been scaling AMD-based clusters and touts sovereign, European AI infrastructure for training, fine-tuning and inference. 

Recent headlines include a partnership with Aker and OpenAI on 'Stargate Norway', a multi-hundred-megawatt site targeting 100,000 NVIDIA GPUs by 2026 – an emblem of Europe’s push for high-density, green AI compute close to cheap, renewable power.

8. RunPod

• HQ: Mt. Laurel, New Jersey, USA
• CEO: Zhen Lu
• Founded: 2022

RunPod has rapidly gained traction by offering cost-effective and flexible GPU cloud computing for AI developers.

The platform is structured to meet diverse needs, providing on-demand GPU clusters through its Secure Cloud and Community Cloud offerings for training and development. 

For inference workloads, its Serverless tier allows users to create custom API endpoints or leverage a suite of pre-existing open-source models.

This tiered approach makes high-performance GPU resources more accessible to a global, remote-first community of AI and ML practitioners.

7. Genesis Cloud

• HQ: Munich, Germany
• CEO: Dr Stefan Schiefer
• Founded: 2018

Genesis Cloud is a European GPU cloud focused on AI and HPC with an emphasis on data sovereignty and sustainability.

The Munich-based provider is building around Nvidia reference architectures and European regulatory expectations, courting enterprises that need performance with predictable jurisdictional control.

It has been active within EU AI initiatives and vendor ecosystems, positioning as a compliant, sovereign alternative to US-centred hyperscale for training and inference workloads across regulated sectors.

6. Vast.ai

• HQ: Los Angeles, USA
• CEO: Jake Cannell
• Founded: 2018

Vast.ai runs a marketplace that aggregates under-utilised GPUs across data centres and professional hosts, exposing them via a single control plane with aggressive spot pricing.

Its proposition is elasticity at low cost: match diverse GPU supply with AI demand while abstracting infrastructure quirks. 

For cost-sensitive training runs or large-scale inference bursts, Vast’s federated model can deliver outsized price-performance – provided workloads tolerate heterogeneity.

The company leans on a community of hosts and automation to keep utilisation high and user set-up lean.

5. Gcore

• HQ: Luxembourg
• CEO: André Reitenbach
• Founded: 2014

Gcore operates as a global edge AI and cloud solutions provider, leveraging an extensive network of over 210 points of presence across six continents.

This vast infrastructure (with a total capacity exceeding 200 Tbps) is designed to deliver low-latency services crucial for real-time AI applications and inference at the edge.

Headquartered in Luxembourg, Gcore provides a comprehensive suite of cloud, network and security solutions, enabling industries from gaming to finance to deploy AI models closer to their end-users for optimal performance.

4. Lambda

• HQ: San Jose, California, USA
• CEO: Stephen Balaban
• Founded: 2012

Lambda is a deep learning infrastructure company providing GPU servers, workstations and extensive cloud services purpose-built for training neural networks. 

Trusted by over 50,000 customers, including major enterprises like Sony and Samsung, research institutions and government agencies, Lambda offers on-demand access to high-performance Nvidia GPUs such as the H100 and H200. 

Its combination of on-premise hardware and a robust, developer-friendly cloud platform makes it a key enabler for organisations building, testing and deploying AI products at scale.

3. Vultr

• HQ: West Palm Beach, Florida, USA
• CEO: J.J. Kardwell (CEO of parent company Constant)
• Founded: 2014

As one of the world's largest independent cloud computing platforms, Vultr offers a comprehensive, full-stack alternative to the dominant hyperscalers. 

Founded with a mission to simplify the cloud, Vultr serves over 1.3 million developers and businesses with a flexible and scalable global infrastructure. 

Its offerings include a wide array of services from bare metal and managed Kubernetes to high-performance cloud GPUs. This extensive product range allows customers to manage both their general-purpose and AI-intensive workloads within a single, cost-effective and easy-to-use platform.

2. Nebius

• HQ: Amsterdam, Netherlands
• CEO: Arkady Volozh
• Founded: 2023

Emerging from the 2024 divestiture of Yandex's Russian assets, Nebius has repositioned itself as a major European-based AI infrastructure powerhouse. 

Led by Yandex Co-Founder Arkady Volozh, the company is building a vertically integrated, full-stack platform for the AI industry, leveraging deep engineering expertise and large-scale GPU clusters across Europe and the US. 

With significant backing from investors including NVIDIA and Accel Partners, Nebius provides the purpose-built infrastructure – from data centres to optimised software – to accelerate AI innovation for start-ups, enterprises and national AI programmes.

1. CoreWeave

• HQ: Roseland, New Jersey, USA
• CEO: Michael Intrator
• Founded: 2017

CoreWeave has established itself as the market leader in specialised cloud infrastructure for AI, delivering unparalleled performance for the most demanding workloads. 

Operating 33 data centres across the US and Europe, the company's close partnership with Nvidia grants it first-to-market access to the latest GPUs, including the GB200 Superchip.

This technological edge has attracted major clients such as Microsoft and OpenAI. 

Following a period of hypergrowth and a successful public listing, CoreWeave continues its massive infrastructure buildout, fuelled by billions of dollars in funding, to meet global demand for high-performance AI compute.