GPU Cloud Provider · Europe
Hyperstack
Hyperstack is a European-based cloud GPU provider designed for high-performance GPU computing, focusing on AI/ML training, rendering, and HPC workflows. It boasts a green infrastructure powered by 100% renewable energy and offers a variety of enterprise-grade features to optimize speed, flexibility, and cost-efficiency.
GPUs
3
Founded
Not stated
Countries
2
Data Centers
2
Uptime SLA
99.9%
Team Size
51-200
GPU Marketplace
Company Profile
Company TypeScale-up
Provider TypeCloud Provider
FoundedNot stated
HeadquartersEurope
Legal EntityHyperstack
Parent CompanyVnet Group (partial stake) / Mirage Group
FundingSeries A/B/C/D
Team Size51-200
Infrastructure
GPU FleetNVIDIA H100 SXM5 80GB, NVIDIA H100 NVL 94GB, NVIDIA A100 SXM4 80GB, NVIDIA RTX A6000 48GB, NVIDIA RTX 4090 24GB, NVIDIA L40S 48GB
Network FabricOptimized for GPU workloads with robust networking features
ConnectivityNot specified
StorageEnterprise-grade storage solutions
Data Center TierTier 3
Bare MetalYes, bare metal GPU instances available alongside virtual instances
AvailabilityGA
EnterpriseStartupResearch
Compute & Deployment
On-DemandYes
Spot / InterruptibleNo
Reserved InstancesYes (monthly and longer-term commitments available)
Bare MetalNo
VM-BasedYes
Container-BasedYes (Docker)
KubernetesYes (self-managed)
Serverless GPUNo
Spin-Up Time2-5 minutes
TerraformYes (official provider)
GPU Hardware
Latest GenH100 SXM, H100 PCIe, L40S
Legacy SupportA100 SXM, A100 PCIe, V100S, RTX A6000
Multi-GPU NodesYes (up to 8x per node)
Max GPUs/Node8
NVLinkYes (NVLink on SXM nodes)
InfiniBandYes (HDR 200Gbps)
PCIe vs SXMBoth PCIe and SXM
HGX PlatformYes (HGX H100 8-GPU)
Pricing Model
Per HourYes (primary billing unit)
Per MinuteNo
SubscriptionYes (monthly plans available)
Reserved DiscountYes (discounts available for longer-term commitments, specific percentages not publicly disclosed)
Spot DiscountNo spot pricing
Public PricingYes
Hidden FeesNone disclosed
Pay-as-you-goYes
Credit SystemYes (prepaid credits system)
Performance & Scaling
Multi-Node TrainingYes (multi-node clusters supported with NCCL)
Elastic ScalingManual only
Auto ScalingNo
InfiniBandYes (InfiniBand available on H100 SXM cluster configurations)
NVSwitchYes (on SXM nodes)
SLA99.9%
Perf IsolationYes (dedicated bare metal)
Noisy NeighborYes (bare metal, no sharing)
Developer Experience
OnboardingSelf-service sign-up with deploy in minutes via web console or REST API; no enterprise onboarding required for standard tiers
FrameworksSupports AI/ML frameworks; exact frameworks not specified
SDK LanguagesPython
CLI ToolingREST API-driven; basic CLI tooling for instance management
JupyterVia SSH port forwarding or self-managed JupyterLab on instances
TemplatesPyTorch, TensorFlow, CUDA
DocumentationComprehensive docs with API reference, quickstart guides, and SDK documentation
API FeaturesCLI, SDK, REST API
Security & Compliance
SecurityNot specified
ComplianceEnterprise-grade security
NVIDIA Cloud PartnerISO 27001 aligned operationsUK-based regulated infrastructure provider
Data Center Locations
Coverage
CountriesUnited Kingdom, Norway
CitiesLondon, Oslo
EuropeNorth America
Compliance Regions
EU Data ResidencyNo EU presence
US Gov CloudNo
India RegionNo
Datacenter Locations
Key Strengths
Competitive H100 pricing compared to hyperscalers
Fast self-service GPU provisioning without waitlists
Multi-region availability across US and Europe
No long-term lock-in with flexible hourly billing
Enterprise-grade NVIDIA hardware accessible to startups
Known Limitations
Smaller ecosystem compared to AWS or GCP
Limited managed ML platform features (no built-in MLOps tooling)
No spot/preemptible instances
Smaller community and third-party integration support
GPU availability can vary by region and model
Additional Information
Support Options
["Enterprise-grade support"]
Community
Limited public community presence; support primarily via documentation and ticket-based system
Core Proposition
High-performance NVIDIA GPU cloud with straightforward on-demand and reserved pricing, targeting AI/ML workloads with fast provisioning and a developer-friendly API.
Payment Methods
Credit CardWire Transfer
Last updated March 2026. Information subject to change.