900-21010-0020-000 NVIDIA H100 NVL 94GB PCIe Accelerator - NEW - 3 YEAR WARRANTY

The NVIDIA® H100 NVL Tensor Core GPU is the most optimized platform for LLM 
Inferences with its high compute density, high memory bandwidth, high energy 
efficiency, and unique NVLink architecture. It also delivers unprecedented acceleration 
to power the world’s highest-performing elastic data centers for AI, data analytics, and 
high-performance computing (HPC) applications. NVIDIA H100 NVL Tensor Core 
technology supports a broad range of math precisions, providing a single accelerator for 
every compute workload. The NVIDIA H100 NVL supports double precision (FP64), singleprecision (FP32), half precision (FP16), 8-bit floating point (FP8), and integer (INT8) 
compute tasks.

The NVIDIA H100 NVL card is a dual-slot 10.5 inch PCI Express Gen5 card based on the 
NVIDIA Hopper™ architecture. It uses a passive heat sink for cooling, which requires 
system airflow to operate the card properly within its thermal limits. The NVIDIA H100 
NVL operates unconstrained up to its maximum thermal design power (TDP) level of 400 
W to accelerate applications that require the fastest computational speed and highest 
data throughput. The NVIDIA H100 NVL debuts the world’s highest PCIe card memory 
bandwidth of nearly 4,000 gigabytes per second (GBps). This speeds time to solution for 
the largest models and most massive data sets.

The NVIDIA H100 NVL card features Multi-Instance GPU (MIG) capability. This can be 
used to partition the GPU into as many as seven hardware isolated GPU instances, 
providing a unified platform that enables elastic data centers to adjust dynamically to 
shifting workload demands. As well as it can allocate the right size of resources from the 
smallest to biggest multi-GPU jobs. NVIDIA H100 NVL versatility means that IT managers 
can maximize the utility of every graphics processing unit (GPU) in their data center.
NVIDIA H100 NVL cards use three NVIDIA® NVLink® bridges. They are the same as the 
one used with NVIDIA H100 PCIe cards. This allows two NVIDIA H100 PCIe cards to be 
connected to deliver 600 GB/s bidirectional bandwidth or 10x the bandwidth of PCIe 
Gen4, to maximize application performance for large workloads