The GPU cloud for your next leap
Train more ambitious models. Bring your AI to production. With dedicated NVIDIA GPUs, high-speed networking, and a cluster built around your business.
Start with 8 GPUs from $19.50/hr · yearly reservation
Behind the teams turning AI into business impact.
01Find your compute
Training, fine-tuning, or inference: every workload has its own demands. Find the capacity that fits your team, your model, and your budget.
See all pricingMemory-rich compute for large models and inference
GPU interconnectNVLink 5 · 1.8 TB/s per GPU
Blackwell Ultra
Frontier model training and FP4 inference
GPU interconnectNVLink 5 · 1.8 TB/s per GPU
New — commissioning
More memory for long-context model serving
Training and fine-tuning with your budget in mind
Prices in USD per 8-GPU node-hour. Rack-scale systems are quoted separately. Subject to capacity and contract terms.
06From research to results
Give your foundation models the capacity they demand. Reserved clusters, dedicated networking, and spare nodes support the infrastructure behind your most ambitious training runs.
Explore B200, GB200, and GB300 →Adapt models to your data and business challenges. Dedicated 8-GPU nodes and persistent storage give your team room to test, fine-tune, and evolve.
Find your H100 or H200 →Bring large models to production with memory-rich GPUs. Serve long-context applications on a dedicated infrastructure foundation you can plan around.
Discover H200 and B300 →02The TensorBay advantage
Every layer of your cluster is designed to give your team more control, more predictability, and more room to build.
F-01
Run directly on dedicated hardware, with the machine’s capacity focused on your workload. Take control of performance at every stage of your project.
F-02
Dedicated nodes, networking, and storage give your team an exclusive foundation to train, experiment, and put AI to work.
F-03
Your cluster goes through 72 hours of load, memory, network, and thermal testing. Review the results and approve delivery with the evidence in hand.
F-04
Your drivers, your kernels, your tools. Work with Slurm, Kubernetes, or SSH and shape the environment around your model’s needs.
F-05
Spare nodes in the same rack and a 15-minute replacement SLA help reduce the impact of hardware failures on your project.
F-06
Published rates, per-minute metering within your reservation, and 20 TB of outbound data included per node each month. Plan with the costs in view.
03Infrastructure behind your ambition
Power, cooling, GPUs, and networking. We design and operate every layer together to give your AI a foundation ready to grow.
1 GW
Contracted power capacity across five countries to support the expansion of our infrastructure.
PUE 1.12
Direct-to-chip liquid cooling in 130 kW racks. Engineered for the demands of intensive AI workloads.
24,000+ GPUs
Capacity expands in six-week cycles, with 72 hours of testing before each handover.
97.4%
Cluster goodput over the last 90 days: the share of capacity translated into useful work.
04Networking and storage
Bring compute, data, and storage together in infrastructure dedicated to your model, from the first training batch to the final checkpoint.
Plan my clusterDedicated networking
Dedicated InfiniBand connects your cluster with bandwidth for distributed training and data exchange between GPUs.
800Gb/s
per InfiniBand XDR link, Quantum-X800 for Blackwell
Storage
Local NVMe, WEKA, and S3-compatible storage to load datasets and save checkpoints at the scale your project needs.
720GB/s
aggregate read throughput per pod, with dedicated WEKA
Data transfer
Free ingress and traffic between nodes, with an included egress allowance to help you plan your operation.
20TB
egress included per node, per month
05Invest in your next move
Choose your reservation term, compare capacity, and plan your investment. Here, the conversation starts with open pricing.
| GPU and memory | Monthly termCapacity reserved on a monthly term | Yearly termLower hourly rate |
|---|---|---|
NVIDIAHGX B300288 GB HBM3e per GPU | Monthly from Quote monthlyUS$ 83.20/h | Yearly from Quote yearlyUS$ 58.50/h |
NVIDIAHGX B200180 GB HBM3e per GPU | Monthly from Quote monthlyUS$ 72.80/h | Yearly from Quote yearlyUS$ 50.70/h |
NVIDIAHGX H200141 GB HBM3e per GPU | Monthly from Quote monthlyUS$ 44.20/h | Yearly from Quote yearlyUS$ 31.20/h |
NVIDIAHGX H10080 GB HBM3 per GPU | Monthly from Quote monthlyUS$ 28.60/h | Yearly from Quote yearlyUS$ 19.50/h |
The hourly rate covers the complete 8-GPU node. Choose a term to request your quote.
Every detail accounted for
See what is included and explore rates for storage, data transfer, and networking.
Talk to a specialistUSD prices per 8-GPU HGX node-hour; rack-scale systems are quoted per rack. Taxes excluded. Managed Slurm and Kubernetes control planes are included. Storage and Direct Connect follow the rates above. Ask about long-term contract options.
Get my proposal07Confidence to move forward
Bring projects to production with dedicated hardware, physical isolation, and access controls. Security is part of the infrastructure, from the first node.
Scope: bare-metal compute, fabric, and managed storage
Certified — all datacenter sites
BAA available on dedicated clusters
EU region with in-country data residency
Published HGX prices are in USD per hour for a complete node with 8 GPUs, within a monthly or yearly reservation. NVL72 systems are quoted per rack. Taxes are not included.
Compare GPU memory, network requirements and your training or inference workload. The hardware section lists H100, H200, B200, B300 and full rack systems. Our team can help size a configuration based on your model, dataset and capacity requirements.
Yes. TensorBay offers dedicated NVIDIA GPU infrastructure for model training, fine-tuning and inference. You can work with Slurm, Kubernetes or SSH and configure the software environment for your workload.
Share the GPU, capacity and reservation term you need, or describe your project. Our team reviews the requirements and prepares a proposal within 48 hours, including configuration and delivery timing. Capacity and final terms are confirmed in the proposal.
08Give your idea room to grow
Bring us your challenge. Our team will help define your GPUs, capacity, and timeline. Get a proposal within 48 hours to launch your project or take it further.
