GPU Marketplace
NVIDIA B200 SXMOn-Demand
Estimates only — rates are collected automatically from public provider pages and may be out of date.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 TypeGPU Cloud
Legal EntityGMI Cloud
Infrastructure
GPU FleetNVIDIA GB200, NVIDIA GB300, NVIDIA H100, NVIDIA H200, NVIDIA B200, flexible deployment across H100, H200, Blackwell and next-generation NVIDIA platforms
designed for enterprise AI workloads at scale.
Compute & Deployment
On-DemandAvailable on-demand for elastic scaling or reserved for predictable long-term workloads.
Reserved InstancesReduce unit GPU costs through reserved capacity and sustained deployments.
Container-Basedfully managed Kubernetes-based GPU clusters
Kubernetesfully managed Kubernetes-based GPU clusters
GPU Hardware
Latest GenFlexible deployment across H100, H200, Blackwell, and next-generation NVIDIA platforms.
Pricing Model
Per Hourfrom $2.00/GPU-hour
Per MinuteCharged by the minute, no upfront cost, pay for what you use
SubscriptionCommitment-Based Savings
Public Pricinghttps://www.gmicloud.ai/en/pricing
Hidden FeesPredictable costs with no hidden fees, sudden throttling, or unexpected constraints.
Pay-as-you-goPay as you go: Charged by the minute, no upfront cost, pay for what you use
Performance & Scaling
Elastic ScalingAvailable on-demand for elastic scaling or reserved for predictable long-term workloads.
Perf IsolationDedicated NVIDIA GPU resources, no shared environments or performance variability.
Noisy NeighborDedicated NVIDIA GPU resources, no shared environments or performance variability.
Developer Experience
OnboardingCluster requests are not processed automatically; after submitting a request, contact support to activate. Pay as you go billing is charged by the minute with no upfront cost. Default OS image is Ubuntu 22.04 x86 64bits.
DocumentationDocumentation includes an index and pages for Managed GPU Clusters: descriptive overview, UI/toolbar details, empty state text, step-by-step Request Cluster workflow (3 steps), list of displayed cluster columns, and a FAQ.
Security & Compliance
Data Center Locations
Coverage
Consistent, competitive pricing across regions with unified billing and full usage transparency.
Compliance Regions
Datacenter Locations
Key Strengths
Claims dedicated NVIDIA GPU infrastructure with predictable pricing and no shared environments
supports on-demand and reserved GPUs across modern NVIDIA platforms
offers commitment-based savings and usage-adaptive pricing
provides unified billing and region-aware pricing.
Known Limitations
Cluster requests are not processed automatically and require contacting support to activate
documentation and pricing mention on-demand and reserved models but do not mention spot instance pricing
model-level pricing is not included on the pricing page.
Additional Information
Core Proposition
Dedicated NVIDIA GPU resources, no shared environments or performance variability.
Payment Methods
Billing methods: Pay as you go or Prepaidsupported by invoicing, consolidated reporting, and region-appropriate payment options
Last updated August 2026. Information subject to change.