Nebius logo

GPU Cloud Provider · Unknown

Nebius

Nebius provides high-performance computing solutions with a focus on AI and machine learning workloads. Their infrastructure features cutting-edge GPU clusters interconnected with InfiniBand networking, offering speeds up to 400 Gbps. Nebius' services are geared towards accelerating complex compute tasks like training and inference in GPU-intensive environments.

GPUs
14
Founded
Unknown
Countries
1

GPU Marketplace

Estimates onlyrates are collected automatically from public provider pages and may be out of date. Prices vary by region, commitment term, and availability, and typically exclude storage, egress, and tax. Confirm current pricing with the provider before purchasing. Last collected .

Company Profile

Provider TypeInfrastructure & network
Legal EntityNebius AI Cloud

Infrastructure

GPU FleetList of GPU instance/platform names as presented on the pricing pages
Network FabricInfiniBand
Connectivity400 Gbps per GPU, up to 3.2 Tbps per node
StorageNVMe, Network SSD
AvailabilityGA (General Availability)

Compute & Deployment

On-DemandOn-demand, GPU-hour
Spot / InterruptiblePreemptible, GPU-hour
Reserved InstancesCommitment discounts
VM-BasedVirtual machines (GPUs, vCPUs, RAM)
Container-BasedManaged Service for Kubernetes® Container orchestrator for AI needs
KubernetesManaged Service for Kubernetes® Container orchestrator for AI needs

GPU Hardware

NVLinkPlatforms labeled NVLink are present (examples: NVIDIA® B300 NVLink, NVIDIA® B200 NVLink, NVIDIA® H200 NVLink, NVIDIA® H100 NVLink).
PCIe vs SXMSX M form-factor platforms are referenced (example identifiers include 'gpu-b300-sxm' and 'gpu-b200-sxm').
HGX PlatformHGX platforms are listed (examples: NVIDIA HGX B300, NVIDIA HGX B200, NVIDIA HGX H200, NVIDIA HGX H100).

Pricing Model

Per Hour1 hour (3600 seconds)
SubscriptionPay up to 35% less than on-demand rates by reserving large-scale clusters for multiple months.
Reserved Discountup to 35% less than on-demand rates
Public Pricinghttps://nebius.com/prices
Hidden FeesAll prices are shown without any applicable taxes, including VAT.
Egress ChargesNetworking: Egress/Ingress traffic Free

Performance & Scaling

Multi-Node TrainingContact us to request large-scale GPU clusters, or sign up in the cloud console to get started right away.
Elastic ScalingBenefit from combining discounts on commitment usage and scalability-friendly, on-demand rates.

Developer Experience

OnboardingSign up in the cloud console to get started right away; to request commitments, sign up for Nebius as a company and contact the sales team.
FrameworksTensorFlow, PyTorch
DocumentationDocumentation includes a documentation index and a Compute pricing article that provides detailed pricing for the Compute service.
API FeaturesCLI, SDK, REST API

Security & Compliance

SecurityLocal compliance standards

Data Center Locations

Coverage

CountriesIsrael

Compliance Regions

Datacenter Locations

Key Strengths

Flexible pricing combining commitment discounts and on-demand rates
up to 35% savings via commitments
preemptible (spot) GPU pricing
convenient payment methods
certain managed services offered for free.

Known Limitations

Many GPU platforms are limited to specific regions (per-GPU-type regional availability)
large-scale GPU clusters must be requested via contact/sales (not self-serve).

Additional Information

Support Options

["Documentation","CLI tools for configuration and management"]

Community

Documentation index available; support ticket link for contacting support.

Core Proposition

Flexible pricing strategy

Payment Methods

Credit cards or bank transfers.
Last updated March 2026. Information subject to change.