Radiant
Radiant is a vertically integrated AI infrastructure company combining powered land, Brookfield capital, NVIDIA-based compute, and the Ori AI Cloud software stack to deliver AI factories for sovereign governments, enterprises, and telecommunications providers under long-term contracts.
- Company typePrivate
- Founded2026
- HeadquartersLondon, United Kingdom
- Headcount101–250
- GTM typeB2B
- OfferingSoftware
What Radiant does
Radiant is a vertically integrated AI infrastructure company that combines powered land, capital, compute, and proprietary software under a single operator. The company was formed in February 2026 through the merger of Brookfield-backed Radiant and London-based Ori Industries, with Brookfield Asset Management valuing the combined entity at $1.3 billion and providing access to up to $100 billion in deployable capital through its AI Infrastructure Fund. Headquartered in London with 101-250 employees, Radiant serves sovereign governments, global enterprises, and telecommunications providers under multi-year contracts, while also operating a self-serve cloud platform for AI builders and developers in markets including the UK, France, Sweden, and Qatar.
The company's core technology stack includes the Ori AI Cloud platform (inherited from Ori Industries), NVIDIA DSX reference design-based AI factories deploying Blackwell, GB200 NVL72, and Rubin architectures, and a powered-land portfolio spanning 5 GW live and 45 GW of renewable generation capacity globally. Technical differentiators include 85% sustained GPU utilization through intelligent scheduling and orchestration, suspend-and-resume capability for idle workloads that can reduce costs by up to 80%, an Inference Delivery Network with cold starts under five seconds, and more than 40 GPU instance configurations. As an NVIDIA Cloud Partner, Radiant integrates from datacenter architecture through software certification, with the proprietary software stack described as dependency-free and built entirely in-house.
Radiant operates a hybrid revenue model combining usage-based self-serve billing (per-minute for GPU instances, per-token for serverless inference, per-minute for dedicated endpoints) with multi-year subscription contracts for sovereign and enterprise deployments where investment-grade customers pay regardless of usage. Brookfield structures these long-term contracts to lock in chip rents across the estimated five-year chip lifecycle. Distribution runs through both direct enterprise field sales and the self-serve Radiant AI Cloud platform, supported by content marketing, developer documentation, a Discord community, and LinkedIn lead generation campaigns.
Radiant firmographics
Firmographics- Name
- Radiant
- Legal name
- Radiant Infrastructure 1 Ltd
- Website
- https://radiant.co
- Company type
- Private
- Founded year
- 2026
- Operating status
- Operating
- Headcount range
- 101–250 employees
- Short description
- Radiant is a vertically integrated AI infrastructure company combining powered land, Brookfield capital, NVIDIA-based compute, and the Ori AI Cloud software stack to deliver AI factories for sovereign governments, enterprises, and telecommunications providers under long-term contracts.
- Ownership category
- akta.pro rank
Radiant industry classification
Industry- Product category
- AI Cloud Infrastructure
- NAICS
- Computing Infrastructure Providers, Data Processing, Web Hosting, and Related Services (5182), Software Publishers (513210), Computer Systems Design and Related Services (54151), Computer Facilities Management Services (541513), Custom Computer Programming Services (541511)
- SIC
- Services-Prepackaged Software (7372), Services-Computer Integrated Systems Design (7373), Services-Computer Programming, Data Processing, Etc. (7370), Services-Facilities Support Management Services (8744)
- akta.pro primary industry
- AI Compute Cloud & GPU-as-a-Service (HDAAAAAK)
- akta.pro secondary industries
- End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management) (HDAEANAA), AI Application Enablement Platforms (Copilot/Agent Frameworks, SDKs) (HDAEANAJ), AI Compute Virtualization & Scheduling (GPU virtualization, cluster schedulers) (HDAAAAAG), AI Datacenter Power, Cooling & Racks (liquid cooling, power delivery) (HDAAAAAH)
Keywords
Where Radiant is headquartered
LocationHeadquarters
- HQ city
- London
- HQ country
- United Kingdom
- HQ region
- Europe
Offices1 record
Markets served
Radiant business model
Business model- GTM type
- B2B
- Offering type
- Software
- Cost components
- Infrastructure, Technology or R&D, Supply Chain, Operations, Personnel, Marketing or Sales
Revenue model
- GPU Compute Instances: On-demand GPU virtual machine instances billed by the minute. Users can choose from 40+ configurations ranging from fractional GPUs to multi-GPU setups, with pre-loaded OS, ML frameworks, and drivers. Pay-as-you-go model for developers and AI researchers.
- Supercomputer Clusters: On-demand supercomputer service providing multi-node GPU clusters for large-scale distributed training. Billed based on usage with dedicated endpoint options for predictable performance and per-minute billing for latency-sensitive workloads.
- Long-term AI Infrastructure Contracts: Long-term contracts with sovereign governments, global enterprises, and telecommunications providers for dedicated AI compute capacity. Brookfield structures contracts to lock in chip rents across the estimated five-year chip lifecycle, targeting investment-grade customers who pay regardless of usage.
- Inference-as-a-Service: Serverless inference endpoints with hardware-abstracted, scale-to-zero, and per-token billing. Dedicated endpoints with exclusive GPU allocation and strict isolation for predictable latency-sensitive workloads billed per minute.
- Managed Kubernetes and Storage: Fully managed, auto-scaling Kubernetes platform and S3-compatible cloud storage with transparent pricing and no hidden ingress or egress fees.
Pricing tiers
| Model | Billing | Price |
|---|---|---|
| Usage-based | Pay-as-you-go | GPU Instances - Pay by the minute with 40+ configurations |
| Usage-based | Pay-as-you-go | Serverless Inference - Per-token billing with scale-to-zero |
| Usage-based | Pay-as-you-go | Dedicated Endpoints - Per-minute billing with strict isolation |
| Subscription | Multi-year contract | Enterprise Long-term Contracts - Commitment Periods with fixed pricing |
Go-to-market motion3 records
Distribution channels3 records
Marketing channels6 records
Radiant product offering
Product offeringCore offering
Radiant is a vertically integrated AI infrastructure company that builds and operates AI factories by combining four pillars: Capital (Brookfield-backed $100B+), Powered Land (5 GW live and 45 GW renewable energy portfolio), Compute (NVIDIA DSX-based GPU clusters with Blackwell, GB200 NVL72, and Rubin), and Software (the proprietary Ori AI Cloud platform). It sells GPU cloud instances, supercomputers, inference endpoints, fine-tuning, model registry, serverless Kubernetes, and storage to sovereign governments, global enterprises, and telecom providers under long-term contracts, alongside a self-serve cloud platform for developers.
Product overview
Radiant is a fully integrated AI infrastructure platform combining four pillars: Software (AI Cloud platform), Compute (GPU infrastructure), Powered Land (data center assets), and Capital (infrastructure financing). The product portfolio centers on the Radiant AI Cloud platform which offers GPU Virtual Machines (40+ configurations), Supercomputers (thousands of NVIDIA GPUs on-demand), Inference Delivery Network (global model routing), Fine-Tuning Studio (PEFT-based model customization), Model Registry (lifecycle management), Serverless Kubernetes (auto-scaling containers), and Cloud Storage (S3-compatible). The platform also includes Sovereign Solutions for governments requiring data localization and Hyperscaler Solutions for large enterprises. Built on NVIDIA DSX reference design with Blackwell and GB200 NVL72 platforms, the vertically integrated stack enables AI factories from concept to deployment.
Differentiator
Problem solved
Functional benefit
Products and services
- Radiant AI Cloud
- GPU Virtual Machines
Quantifiable outcome
- 85% GPU utilization vs 50-60% industry average
- +4 more outcomes
Companies that use Radiant
Customer profileNamed customers4 records
Segments4 records
Ideal customer profiles4 records
Radiant technology and API
TechnologyTechnology focussed Yes
API detail
- Has API
- Yes
- API docs
- API detail
Core technology
AI maturity
App detail
Integration2 records
AI capability6 records
Feature6 records
Radiant partnerships and signals
Strategic signalPartnerships
Four partnerships are on record, tiered core and major.
- NVIDIAcoreRadiant is an NVIDIA Cloud Partner utilizing NVIDIA DSX reference design for AI factory infrastructure. NVIDIA contributes to the initial capital of Brookfield's AI fund and supplies chips to Radiant. The partnership covers datacenter architecture to software certification, deploying Blackwell, GB200 NVL72, and Rubin platforms.
- Ori IndustriescoreOri Industries merged into Radiant on February 24, 2026, combining Ori's distributed AI infrastructure software platform with Radiant's global infrastructure capabilities. The merger created one of the world's first vertically integrated AI infrastructure companies.
- Bloom EnergymajorBrookfield plans to invest up to $5 billion in Bloom Energy for fuel cells at data centers to power AI infrastructure. Bloom Energy will provide power generation solutions supporting Radiant's powered land strategy.
- IntelmajorBrookfield struck a $30 billion partnership with Intel to expand chip plants in Arizona, which supports the broader AI infrastructure ecosystem including Radiant's supply chain.
Scale indicators7 records
Recent moves7 records
Expansion highlights6 records
Radiant competitors and assessment
Company assessmentDirect peers
- Crusoe Energy: Crusoe builds large-scale AI datacenters powered by stranded and renewable energy, combining powered land with GPU compute. Like Radiant, its pitch centers on pairing energy assets with NVIDIA-grade AI infrastructure for hyperscale customers.
- Nebius: Nebius operates an AI-native cloud platform with NVIDIA GPU instances, inference, and fine-tuning services. It is a direct competitor to Radiant in serving sovereign and enterprise AI workloads, and pursues a similar vertically integrated AI infrastructure thesis.
- CoreWeave: CoreWeave is a GPU-as-a-service cloud built specifically for AI workloads, offering NVIDIA H100/Blackwell instances, Kubernetes, and inference services. It is the closest direct competitor to Radiant AI Cloud's self-serve GPU platform and serves similar AI builder and enterprise customers.
- Lambda Labs: Lambda provides NVIDIA GPU cloud instances, on-prem GPU clusters, and inference services to AI developers and enterprises. It competes directly with Radiant AI Cloud on per-minute GPU pricing, multi-GPU configurations, and MLOps tooling.
- Vultr: Vultr provides global cloud compute including dedicated and virtualized NVIDIA GPU instances aimed at developers and AI startups. It overlaps with Radiant's self-serve AI Cloud on per-minute GPU billing and global footprint for AI builders.
- Together AI: Together AI runs a GPU cloud optimized for open-source and frontier model training and inference, offering per-token and per-hour GPU pricing. Its inference and fine-tuning stack overlaps directly with Radiant's Inference Delivery Network and Fine-Tuning Studio.
Emerging players
- Nscale: Nscale is building sovereign AI infrastructure with powered land and GPU clusters across the UK and Europe. It pursues a similar sovereign-AI GTM motion to Radiant and competes for the same government and regulated-industry customers, though at smaller scale.
Others
- Equinix: Equinix is a global colocation and interconnection provider that hosts AI infrastructure deployments. While not a direct GPU cloud competitor, it is a relevant ecosystem peer given Radiant's powered-land AI factory buildout that may leverage carrier-neutral facilities.
Broad incumbents
- Amazon Web Services (AWS): AWS offers EC2 GPU instances, SageMaker, Bedrock inference, and EKS for AI workloads. As the largest hyperscaler, it competes broadly with Radiant across both sovereign/enterprise deals and the self-serve AI builder segment.
- Microsoft Azure: Azure provides ND-series GPU VMs, Azure ML, Azure OpenAI Service, and Kubernetes for AI training and inference. It is a broad incumbent offering overlapping capabilities with Radiant's full AI Cloud stack to enterprise and sovereign customers.
Market position
Strengths5 records
Weaknesses5 records
Competitive moat6 records
Key risks6 records
Key highlights7 records
Customer concentration
Radiant social profiles
Digital presenceRadiant compliance and trust
Trust signalCompliance1 record
Radiant financial estimates
Financial estimateRevenue estimate
Valuation estimate
Radiant leadership team
Management profileNumber of profiles
Profiles6 records
Radiant subsidiaries and ownership
Company hierarchySubsidiaries1 record
Radiant funding detail
Funding detailFunding overview
Funding rounds
Investors
Funding detail is available on the Subscription and Enterprise plan.Contact sales →
Radiant M&A and investment
M&A and investmentM&A1 record
Investments
M&A and investment is available on the Subscription and Enterprise plan.Contact sales →
Frequently asked questions about Radiant
What does Radiant do?
Radiant is a vertically integrated AI infrastructure company that builds and operates AI factories by combining four pillars: Capital (Brookfield-backed $100B+), Powered Land (5 GW live and 45 GW renewable energy portfolio), Compute (NVIDIA DSX-based GPU clusters with Blackwell, GB200 NVL72, and Rubin), and Software (the proprietary Ori AI Cloud platform). It sells GPU cloud instances, supercomputers, inference endpoints, fine-tuning, model registry, serverless Kubernetes, and storage to sovereign governments, global enterprises, and telecom providers under long-term contracts, alongside a self-serve cloud platform for developers.
Is Radiant a public or private company?
Radiant is a private company. It is classified as private equity controlled and is currently operating.
When was Radiant founded?
Radiant was founded in 2026. It employs 101 to 250 people.
Where is Radiant based?
Radiant is headquartered in London, United Kingdom, in the Europe region.
How does Radiant make money?
Five revenue lines are on record. GPU Compute Instances are the primary driver. The others are supercomputer Clusters, long-term AI Infrastructure Contracts, inference-as-a-Service and managed Kubernetes and Storage.
Who are Radiant's main competitors?
Direct peers on record are Crusoe Energy, Nebius, CoreWeave, Lambda Labs, Vultr and Together AI. Nscale is listed as an emerging player. Equinix is listed as an others. Broad incumbents are Amazon Web Services (AWS) and Microsoft Azure.
Does Radiant have an API?
Yes. Radiant provides a public OGC REST API (v1.0.0) for creating and managing cloud resources programmatically. The API supports authentication via Bearer Token (API Key) using the Authorization header. It provides endpoints for managing Virtual Machines, Supercomputers, Kubernetes clusters, Inference Endpoints, SSH keys, and health/version checks. API tokens are created in Settings under the API Tokens tab. Developer documentation is at docs.radiant.co/api/overview.
What industry is Radiant in?
Radiant's product category is AI Cloud Infrastructure. Its primary akta.pro industry code is HDAAAAAK, AI Compute Cloud & GPU-as-a-Service, with a secondary code of HDAEANAA, End-to-End Enterprise AI Platforms (MLOps & Model Lifecycle Management). Its NAICS code is 5182 and its SIC code is 7372.