Rate Limits
To ensure service availability and performance, we have global and endpoint-specific rate limits.
Global rate limits apply to API requests to all endpoints by a developer application.
Endpoint-specifc rate limits apply to API requests to a certain endpoint or endpoint group. Rate limits for different endpoints are independent. For example, if the rate limits for /campaign/get endpoint are reached for a developer app, the developer app can still make requests to /ad/get/ endpoint.
Global rate limits
We set four different call limit levels for each developer application, and each level includes three different types, QPS Limit (Queries-Per-Second Rate Limiting), QPM Limit (Queries-Per-Minute Rate Limiting) and QPD Limit (Queries-Per-Day Rate Limiting).
Once the rate limit is met, the server returns "code": 4034, which means your request was throttled. You will need to suspend calls to that interface for a period of time and resume calls after the suspension has been lifted. The suspension time depends on the type of rate limit set. For QPM limit, you need to wait 5 minutes before you can make API requests again. For QPD limit, you need to wait until the next day (UTC+0 time) to make API requests again. Note that the QPD limit resets at 00:00:00 UTC+0 time every day.
Rate limit levels
| Level | QPS | QPM | QPD |
|---|---|---|---|
| Basic | 10 | 600 | 864,000 |
| Advanced | 20 | 1,200 | 1,728,000 |
| Premium | 30 | 1,800 | 2,592,000 |