Immediate access to NVIDIA H200 Tensor Core GPUs

From a single GPU to thousands. Choose on-demand, pay-as-you-go options or cost-effective reserved plans.

The most powerful GPU for your AI and HPC workloads

Memory

The first GPU with 141 GB of HBM3e memory — nearly double the memory capacity of NVIDIA H100 SXM.

Performance

The increased memory bandwidth of 4.8 TB/s allows for better utilization of processing power, making it the new preferred GPU for LLMs.

Data access speed

With the shared filesystem, you can achieve up to 20GB/s read speeds from one node, which is crucial for training and inference. This is 6 times more than what was available with the NVIDIA H100.

Prices for NVIDIA H200 Tensor Core GPU

Contact our sales team to get the best pricing and a free proof-of-concept stage for your AI project.

$2.30 GPU/h

As low as $2.30 per 1 GPU/hour, the NVIDIA H200 with Intel Sapphire Rapids is available now.

$2.59 GPU/h

Price for 1 GPU/hour with a minimum of 1-month reserve commitment to a minimum of 8 GPUs.

NVIDIA H200 vs NVIDIA H100

According to an NVIDIA research, NVIDIA H200 shows next-level performance compared to NVIDIA H100 for popular LLMs: Llama2 13B, Llama2 70B and GPT-3 175B.

NVIDIA H200 SXM offers significant enhancements over the NVIDIA H100 SXM, delivering up to 45% better performance in generative AI and HPC tasks.

NVIDIA GPU clusters with InfiniBand

We provide NVIDIA GPU clusters with InfiniBand interconnect.
InfiniBand is a technology that facilitates seamless connection of servers equipped with NVIDIA GPUs within a cluster. Leverage multi-node interconnection with up to 3.2 Tbit/s.

How the committed reserve works

Place an order now to lock in the price for your future GPU needs.

Our sales team will contact you soon to confirm the dates and provide the contract, which must be signed at least one month before your reserve starts. You can adjust the reserve dates anytime up to a month before the start.