NVIDIA developed the A100 and H100 graphics-processing units used widely in artificial-intelligence computing.
NVIDIA was founded in 1993 by Jensen Huang, Chris Malachowsky, and Curtis Priem. It first became famous for graphics-processing units used in personal computers and video games. GPUs perform many calculations in parallel, making them useful for rendering images as well as training and running machine-learning models.
The A100, based on NVIDIA’s Ampere architecture, and the H100, based on Hopper, became important products for data centers. NVIDIA also supplies CUDA software, a programming platform that lets developers use its GPUs for scientific computing and artificial intelligence.
This combination of specialized hardware and software helped drive a dramatic increase in NVIDIA’s market capitalization. The company is sometimes confused with a computer manufacturer, but it primarily designs chips and platforms while relying on semiconductor foundries for fabrication. Market value reflects expected future demand as well as current sales.