Skip to main content

NVIDIA H100

Test now in the MEGWARE Benchmark Center

An order-of-magnitude leap for accelerated computing.

The NVIDIA H100 Tensor Core GPU delivers unprecedented acceleration at every scale for AI, data analytics, and high-performance computing (HPC) to tackle the world’s toughest computing challenges. With NVIDIA® NVLink® Switch System, up to 256 H100s can be connected to accelerate exascale workloads, along with a dedicated Transformer Engine to solve trillion-parameter language models. H100’s combined technology innovations can speed up large language models by an incredible 30X over the previous generation to deliver industry-leading conversational AI.

GET YOUR FREE TEST DRIVE


Contact us and we put together the best test configuration for your requirements.

Technical Data

FP6426 TF
FP64 Tensor Core51 TF
FP3251 TF
TF32 Tensor Core756 TF
BFLOAT16 Tensor Core1,513 TF
FP16 Tensor Core1,513 TF
INT8 Tensor Core3,026 TOPS
GPU memory80 GB
GPU memory bandwidth2 TB/s
Max thermal design power (TDP)300 bis 350 W (configurable)
Multi-Instance-GPUsUp to 7 MIGs @ 10 GB each

 

NVIDIA H100 PCIe

NVIDIA H100

PCIe

Form factor

PCIe

Dual-slot air-cooled

InterconnectNVLINK: 600GB/s PCIe Gen5: 128 GB/s

 

Technical Data

FP6434 TF
FP64 Tensor Core67 TF
FP3267 TF
TF32 Tensor Core989 TF
BFLOAT16 Tensor Core1,979 TF
FP16 Tensor Core1,979 TF
INT8 Tensor Core3,958 TOPS
GPU memotry80 GB
GPU memory bandwidth3.35 TB/s
Max thermal design power (TDP)Up to 700 W (configurable)
Multi Instance GPUsUp to 7 MIGs @ 10 GB each

 

NVIDIA H100 SXM

NVIDIA H100

SXM

Form factor

SXM

InterconnectNVLINK: 900GB/s PCIe Gen5: 128 GB/s