What does TOPS mean in AI?

TOPS (Trillions or Tera Operations per Second) is a classic example of a simplifying metric.  It tells you how many computing operations an AI chip can handle in one second — how many basic math problems a chip can solve in that very short period of time.

While TOPS doesn’t differentiate between the types or quality of operations a chip can process, if one AI chip offers 5 TOPS and another option offers 10 TOPS, you may correctly assume that the second is twice as fast as the first.

Though some people dislike the use of abstract performance metrics when evaluating computing capabilities, customers tend to prefer simple, seemingly understandable distillations to the alternative.