NVIDIA makes the H100 GPU family, a product line that helped drive its rise among the world's most valuable companies.
The H100 is a data-center graphics processing unit based on NVIDIA's Hopper architecture. It is designed for demanding artificial-intelligence and high-performance-computing workloads, including training and running large machine-learning models.
NVIDIA's market capitalization expanded dramatically as cloud providers and technology companies invested in accelerated computing. Its CUDA software platform also became an important part of the company's competitive position because developers use it to program NVIDIA GPUs for scientific and AI applications.
The H100 is sometimes described as a conventional graphics card, but its main importance is as a data-center accelerator. NVIDIA also sells gaming GPUs under the GeForce brand. Market capitalization reflects investor expectations about the whole company, not the sales of one chip or product family alone.