Rate limits and usage

Use this guide to identify what limits your usage, distinguish 429 from 402 responses, and keep production executions running.

Execution counts do not limit your account

Runtype does not cap the number of Flow executions that you can run. No plan has a daily or monthly execution pool, and no plan charges execution overage fees. AI model calls and built-in tools determine your spend. Control that spend with prepaid credit, auto-topup, and an optional spend cap.

Your billing state does not change the execution count. Accounts with and without a payment method on file have no execution-count cap.

The execution count remains unlimited in both billing states:

These controls apply to Flow executions, not to individual AI model calls. Each model provider enforces separate limits. For a comparison of platform keys and BYOK, see Platform keys vs. bring your own key (BYOK).

What can return a 429

Some plans have a burst rate that limits the number of Flow executions that you can start per second. If you exceed that rate, the API returns 429. The Retry-After header gives you the retry interval. Plans without a burst rate skip this check.

For an execution request, a burst-rate check is the only Runtype limit that returns 429. Execution count does not cause a 429 response.

What can return a 402

Spending controls return 402, not 429, in these cases:

• Your prepaid credit is exhausted and auto-topup is not configured or fails.
• Your free trial has ended and you have no payment method on file.
• Your month-to-date spend reaches your monthly spend cap.

Review these conditions in Settings > Billing. Update your payment method, auto-topup settings, spend cap, or plan in Settings > Billing.

Platform keys and BYOK

Your key setup determines which AI provider limits apply to each model request.

Platform keys use shared Runtype capacity. Provider throttling can affect response times during peak usage.

BYOK uses your provider account’s limits and bills you directly through that provider. Choose BYOK for production workloads or when you need more consistent response times. To connect BYOK, follow Connecting AI model providers.

Check your usage

Use the dashboard and API responses to monitor usage.

• Review today’s execution count in the dashboard. The count is informational and does not show a limit fraction.
• Use Usage data to review spend by model and tool.
• Review credit, auto-topup, and your spend cap in Settings > Billing.

When a burst rate applies, execution responses include RateLimit-Limit, RateLimit-Remaining, and RateLimit-Reset. Every account includes X-Quota-Policy: unlimited and no numeric X-Quota-* headers because execution counts have no cap.

Reduce cost and latency

Use these practices to reduce cost and latency:

• Cache repeated results in Records when you do not need to run the same Flow again.
• For large workloads, lower concurrency and add delay steps when your plan has a burst rate.
• Add retry logic that respects the Retry-After header on 429 responses.
• Choose a smaller model for high-volume steps. Spend, not execution count, determines platform-key cost.

Troubleshoot production limits

If a production request hits a limit, follow these steps:

1. Check whether the response is 429 (rate) or 402 (spend). Use the matching fix.
2. If your plan has a burst rate, add delays between high-frequency steps in your Flows.
3. If platform-key throughput limits the request, switch to BYOK.
4. Review your credit balance, auto-topup, and spend cap in Settings > Billing.
5. Upgrade your plan when you need features such as BYOK, faster schedules, or longer log retention.

For model configuration, follow the provider setup steps in Settings > Models.

Next steps

Continue with these guides:

• Daily executions: interpret the dashboard execution count.
• Managing AI models: choose models for your Flows.
• Billing and plans: review billing controls and plan features.