Skip to main content
Rase AI Intelligence Newsletter

AI infrastructure Product

NVIDIA’s H100 brought Hopper to data center AI workloads

The NVIDIA H100 is a data center accelerator built on the Hopper architecture for AI training, inference and high-performance computing. [1]

What it does

NVIDIA introduced a Transformer Engine with Hopper to run some operations in lower precision while using other precision formats where appropriate. [1]

Why it matters

Precision and memory use affect how much computing a model run requires. The Transformer Engine gives teams another hardware feature to evaluate for their workloads. [1]

Key dates

  1. NVIDIA announces Hopper and the H100 [1]

Sources and verification

Catalog evidence checked How we verify records

  1. Primary source NVIDIA Announces Hopper Architecture, the Next Generation of Accelerated Computing | NVIDIA Newsroom
  2. Independent source Nvidia Reveals Hopper H100 GPU With 80 Billion Transistors | Tom's Hardware

Related profiles

AI infrastructure

Azure Cobalt 200 virtual machines

Microsoft Azure

The Arm-based CPU platform supports databases, applications and tool execution around AI models.

High confidenceEvidence checked

AI infrastructure

Sarvam Token Factory

Sarvam AI

It exposes hosted model inference through supported developer interfaces.

High confidenceEvidence checked

AI infrastructure

NVIDIA Rubin HPC

NVIDIA

It is designed to accelerate high-performance computing workloads in compatible systems.

High confidenceEvidence checked