Reference · GlossaryToken quotaA **token quota** caps how many model tokens a user, tenant, or job may consume in a period.On this pageWhen to useWhen notExample#When to useMulti-tenant SaaS, free tiers, and batch jobs with budget caps.#When notWithout alerting — silent throttling confuses users.#Example“Pro plan: 500k tokens/day; hard stop at 110% with upgrade prompt.”Learn by doing → Learn by doing →Related termsRate limitCost per tokenAPI rate limit← All terms