Metric & Benchmark

Throughput

Tokens or requests processed per second by an inference system.

Definition

Throughput measures system-level capacity — total useful work per unit time. For LLMs it is typically reported in tokens/second across the whole batch, distinct from per-user latency. Maximising throughput often trades off against per-request latency.

Common use cases

  • Capacity planning
  • Cost modelling
  • Serving design

Related terms

    Throughput — AI Glossary | Railwail