Supermicro NVIDIA HGX H100/H200 8-GPU Systems - JumpStart Program

NEW: Supermicro NVIDIA HGX H100/H200 8-GPU Systems now available for testing and validation.

Our JumpStart program offers convenient remote sessions to fully configured systems. It allows you to validate, test, and benchmark your workloads with a few simple steps.  
 
Supermicro NVIDIA HGX H100/H200 Systems are the building blocks of the world’s most powerful AI clusters for large-scale training and inference, equipped with improved thermals, reduced acoustics, and cluster-ready connectivity. 
 
These systems power 8 NVIDIA HGX H100/H200 GPUs, interconnected with 900GB/s GPU-GPU NVLink for a combined pool of 640GB HBM3 memory. Supermicro systems with HGX H100/H200 GPUs bring PCIe 5.0 connectivity, fourth-generation NVLink GPU-GPU interconnect and 800Gb/s networking for scale-out, and the new NVIDIA ConnectX®-7 and BlueField®-3 cards empowering GPUDirect RDMA and Storage with NVIDIA Magnum IO and NVIDIA AI Enterprise software. 

Please fill in the information below to sign up to test the systems.

Supermicro reserves the right to revise, update, cancel or terminate the JumpStart program without prior notice and may do so at its sole discretion. In addition, Supermicro reserves the right to disqualify any applicant from the JumpStart program. In recognition of demand fluctuations and service availability, access to the JumpStart program is based on a limited time offer, the term of which is subject to Supermicro’s sole discretion. Supermicro may not be held liable based on any lack of availability of the program to the participants and it is not liable for unfulfilled requests from participants.