Peak rate
A peak rate is a reference maximum for a specified resource, operation and operating assumption. It is the denominator of many relative-throughput metrics.
How it works
Theoretical peaks derive from hardware capabilities and clocks. Profiler sustained-peak conventions can use more specific resource definitions. Precision, sparsity, clock state and active-versus-elapsed normalization all affect what a percentage means.
Work through an example
A kernel delivering 40 TFLOP/s is at 40% of an illustrative 100 TFLOP/s applicable ceiling. Dividing by a 200 TFLOP/s sparse ceiling would answer a different question.
Where the term can mislead
A low fraction of peak can result from dependencies or a mismatched denominator. It does not automatically imply poor implementation.
Connect the idea
Take it into practice
Continue with the existing concept, apply it to an interview problem, or follow a course lesson. Access requirements are shown before you open a page.
New to GPUs? Start with the free execution-model concept.
Sources and scope
Examples here are teaching calculations, not measured GPU performance. Consult the linked documentation for the exact device, instruction and tool version. Source review: 25 October 2026.