Feed Ask Research Search

Feed Ask Research Search

Home Search Ask Research Alerts

A100 · podgraph

A

A100

Product · mentioned 4 times

Data-center GPU from NVIDIA, based on the Ampere architecture and used for AI training.

Tap a name to go there · tap a line to see why they are linked

Made by 1

Latest news

Minus One· 21 May
Amit Jain says the A100, built on the Ampere architecture, was the first AI training chip but was never designed for training and was extremely inefficient compared with today's chips.