The useful result
Send a large batch of items to a rate-limited API without 429 errors.
- Send a large batch of items to a rate-limited API without 429 errors.
A workflow that works for five rows fails at fifty with “The service is receiving too many requests from you”. That is a 429 rate-limit response: the service allows only so many calls in a given time. The cure is not to retry harder, it is to send slower.
n8n offers three tools for this, and its rate-limit guide describes them: the Batching option on the HTTP Request node, a Loop Over Items plus Wait pattern, and Retry On Fail with a pause between tries. This lesson shows how to choose and test one.
The examples below are illustrative exercises, not recorded results from any tool. Sources were checked on 24 October 2026; menus, limits and prices change, so confirm details on the linked pages.
Before you start
- An n8n workflow that calls an API for many items
- The API’s documented rate limit (requests per second or minute)
- A test set of about 20 items
Review evidence
Documentation-based review · 2026-10-24
Primary documentation and official guidance checked on 24 October 2026. Examples, checks and exercises are original illustrations, not recorded tool tests. Cover artwork is a conceptual illustration.
Find the limit in the API’s own documentation
Look up the documented limit, such as one request per second or sixty per minute. Work out a safe delay that stays under it, and leave a margin. Do not guess from the error message, because the limit may count per key, per account or per endpoint.
Check your result
You can state the limit and the delay you will use in one sentence.
Choose one slowing method and set it
In the HTTP Request node, add the Batching option and set Items per Batch and a Batch Interval in milliseconds. Or place Loop Over Items before the call and a Wait node after it, looping back. Either way the requests are spaced out instead of fired together.
Check your result
The workflow sends items in small groups with a pause between them.
Add Retry On Fail, then test and check for duplicates
In the node’s Settings, turn on Retry On Fail and set Wait Between Tries above the rate-limit window. Run your 20-item test and watch for 429 errors and for duplicates, because a retried request that creates a record can create it twice.
Check your result
The 20-item test finishes with no 429 errors and no duplicate records.
When it goes wrong
A very short interval may still trip the limit when other workflows share the same key. Retrying a request that creates data without an identifier can create duplicates. Paid AI APIs may also bill failed or repeated calls, so check the provider’s pricing rules.
Make it your own
Send 20 test items to a free or sandbox API using your chosen method. Record the batch size, the interval, the total run time and the number of errors.
Useful next resources
Sources and review
Last checked Oct 24, 2026. Features and account limits can change. Verify important results against your original material.
Did this help you complete your task?
Tell us what needs improving. No account or private output required.
Report incorrect information or a broken linkGenius Forges Editorial
Practical AI lessons with checked sources, labelled examples and human verification.
