On demand and reserved GPU power for your deep learning workflows

Train your models with NVIDIA H100 and A100 with no disruption in speed.

NVIDIA H100 GPU
The flagship GPU for data centers and the main driving force behind the world's AI infrastructure. Built on the most recent NVIDIA Hopper architecture and available in both PCIe and SXM configurations.

Exceptional performance through the cloud built for deep learning

We provide access to the widest range of NVIDIA GPUs in the industry, allowing you to scale according to the demands of your workloads. Our infrastructure, native to Kubernetes, ensures rapid deployment times, dynamic auto-scaling, and a modern networking architecture that grows with your needs.

Optimize your deep learning workloads

Every model is unique, and so are its computational needs. With the most extensive selection of GPUs in the industry, you can train, fine-tune, and deploy models more quickly and efficiently.

Experience bare metal performance through Kubernetes

Eliminate hypervisors from your stack by deploying containerized workloads. We enable you to enjoy the advantages of bare-metal performance without the hassle of managing infrastructure.

Comprehensive expertise in Machine Learning

Machine Learning is at the heart of what we do, and our infrastructure reflects this. Whether you're training or deploying models, our cloud service is designed to minimize your setup time and enhance performance.

Designed for deep learning workflows

Get ready for machine learning with a pre-configured environment that allows you to start training instantly. Enjoy the convenience of one-click Jupyter access and effortlessly connect to NVIDIA GPU instances from your browser. Plus, benefit from pre-installed popular ML frameworks like Ubuntu, TensorFlow, PyTorch, CUDA, and cuDNN, all readily available.

Harness the power of NVIDIA A100 and H100 at incredible prices.

Experience an order-of-magnitude leap for accelerated computing with CoreCluster. We offer on-demand access to NVIDIA A100 and H100 Tensor Core GPUs, providing unprecedented performance, scalability, and security for every workload. Whether you're training complex models, running high-performance applications, or exploring new frontiers in AI, our GPUs provide the power and speed you need.

Market-leading pricing
ON-DEmand pricing

Access high-performance GPUs when you need them and pay for the time you use them.

We offer some of the best prices and value for NVIDIA H100 and A100 clusters on the market.

Model
PLATFORM
GPUs
VRAM (GB) per GPU
vCPUs per GPU
RAM (GB) per GPU
Storage (GB) per GPU
On-demand Price per Hour (3 month term)

NVIDIA H100

PCIe

1

80

48

256

500

$4.60

NVIDIA H100

PCIe

2

80

48

256

500

$9.13

NVIDIA H100

PCIe

4

80

48

256

500

$18.33

NVIDIA H100

PCIe

8

80

48

256

500

$36.68

NVIDIA H100

HGX

1

80

48

256

500

$4.90

NVIDIA H100

HGX

2

80

48

256

500

$9.73

NVIDIA H100

HGX

4

80

48

256

500

$19.50

NVIDIA H100

HGX

8

80

48

256

500

$39.08

NVIDIA A100

PCIe

1

80

48

256

500

$2.80

NVIDIA A100

PCIe

2

80

48

256

500

$5.53

NVIDIA A100

PCIe

4

80

48

256

500

$11.10

NVIDIA A100

PCIe

8

80

48

256

500

$22.28

NVIDIA A100

HGX

1

80

48

256

500

$3.30

NVIDIA A100

HGX

2

80

48

256

500

$6.53

NVIDIA A100

HGX

4

80

48

256

500

$13.10

NVIDIA A100

HGX

8

80

48

256

500

$26.28

Join the accelerator program, start for free.

Are you a start up with fewer than ten people? Join our accelerator program to get free access to NVIDIA H100 Tensor Core GPU.