Traffic controls

Rate limits answer different failure modes

RPM controls request count, TPM controls token volume, concurrency limits simultaneous work, and budgets cap spend over time.

Planned recharge methods: Alipay and USDT. Collection is not enabled yet.

Use layered limits

  • Project budget for the environment
  • Key budget for the integration
  • RPM for request bursts
  • TPM for token-heavy traffic
  • Concurrency for resource pressure
  • Application loop limits for agents

Return a useful user experience

When a limit is reached, surface a clear message and a retry time where applicable. Do not silently switch to an untested model just to keep a request moving.

Common questions

Is RPM enough for AI agents?

No. An agent can produce large requests or long outputs within a modest request count. Combine RPM with TPM, concurrency, budgets and loop limits.

Where should limits be set?

Set broad limits at project level and narrower limits on each key according to the integration's purpose.

Related resources