Infrastructure atlas
NVIDIA/Hopper

H100 Tensor Core

A Hopper data-center GPU with Tensor Cores, Transformer Engine FP8, NVLink, and MIG partitioning.

GPUtraininginferenceHPCCUDA
H100 Tensor Core hardware

01 / Memory

80 GB HBM3

SXM5 configuration

02 / Bandwidth

3.35 TB/s

GPU memory bandwidth

03 / Power

Up to 700 W

Configurable SXM TDP

Precision & peak compute

PrecisionPublished peakBasis
FP8 Tensor3,958 TFLOPSSXM, with sparsity
FP16 / BF16 Tensor1,979 TFLOPSSXM, with sparsity
INT8 Tensor3,958 TOPSSXM, with sparsity

Peak FLOPS and TOPS are theoretical hardware figures. Sparse and dense modes, chip and system totals, and form factors are not interchangeable.

NVIDIA H100 specifications

Technical specification

Silicon & memory
Architecture
Hopper (GH100)
Process
TSMC 4N
Memory
80 GB HBM3 · 5,120-bit
FP8 Tensor
3,958 TFLOPS with sparsity
FP16 / BF16 Tensor
1,979 TFLOPS with sparsity
System & interconnect
Form factors
SXM5 or PCIe
Scale-up
NVLink 4 · up to 900 GB/s bidirectional
Partitioning
Up to 7 MIG instances
Software
CUDA, cuDNN, TensorRT, NCCL, MIG