GPU Cloud Provider
IREN Cloud
Company Profile
Provider TypeAI Cloud
Legal EntityIREN
Infrastructure
GPU FleetNVIDIA H100, NVIDIA H200, NVIDIA B200, NVIDIA B300, NVIDIA GB300 NVL72
Bare MetalThe bare metal GPU servers provide environment control with root access and no restrictions.
sensitive, enterprise and regulated workloads
Compute & Deployment
Bare Metalbare metal GPU servers
Container-BasedContainers (including Docker and Apptainer)
GPU Hardware
Latest GenNVIDIA H100NVIDIA H200NVIDIA B200NVIDIA B300NVIDIA GB300 NVL72
InfiniBandNVIDIA InfiniBand with 3.2TB/s bandwidth
HGX PlatformNVIDIA HGX B300
Pricing Model
Performance & Scaling
Multi-Node TrainingThis high-speed interconnect is essential for distributed AI model training, parallel processing across GPUs, and low latency AI/ML training.
Elastic ScalingLeverage AI infrastructure designed to scale and grow with your business demands.
InfiniBandfull non-blocking 3.2TB/s InfiniBand connectivity
Perf IsolationAccess secure, isolated environments for different users or teams for data privacy and resource allocation.
Noisy NeighborYes, our isolated environments use private networking and strict tenant separation, making them suitable for sensitive, enterprise and regulated workloads.
Developer Experience
OnboardingContact sales / Talk to us (sales-led onboarding)
DocumentationFrequently Asked Questions (FAQ) and a resources section containing articles, blogs, case studies, and videos; per-GPU SPECS are listed on the page.
Security & Compliance
NVIDIA Exemplar Cloud status on NVIDIA HGX B300case study customer Hume"trusted delivery partners" supported by tier 1 engineering firms and vendor partners24/7 in-house support mention.
Data Center Locations
Coverage
Compliance Regions
Datacenter Locations
Key Strengths
Built to NVIDIA reference architecture with full non-blocking 3.2TB/s InfiniBand
owned and operated purpose-built data centers (vertical integration)
24/7 in-house support
secure isolated environments with private networking and strict tenant separation
flexible storage including on-node and WEKA.IO.
Additional Information
Community
Articles, Blogs, Case Studies, Videos (resources / latest updates content)
Core Proposition
Built to NVIDIA reference architecture with full non-blocking 3.2TB/s InfiniBand connectivity
Notable Customers
Hume
Last updated August 2026. Information subject to change.