Every Airtable API limit, in one table
Last verified
Airtable limits API use in two ways. The rate limit applies on every plan: 5 requests per second per base, plus 50 requests per second for all traffic from one user's (or service account's) personal access tokens. Exceed either and you get HTTP 429 and must wait 30 seconds before requests succeed again. The monthly call limit applies on the Free and Team plans only: 1,000 API calls per workspace per month on Free and 100,000 on Team. Business and Enterprise Scale have no monthly cap, and only the per-base rate limit applies.
Write throughput comes from the two limits together. Create, update and delete calls take up to 10 records each, so the documented ceiling is about 50 records per second per base. Reads return at most 100 records per page. Plan limits also cap how much data a base can hold: 1,000 records per base on Free, 50,000 on Team and 125,000 on Business, with matching attachment storage caps.
Values come from Airtable's developer reference and support center, verified November 12, 2026. Airtable says it may change the enforced limits or tier them by plan. Plan figures here are as of November 2026.
Official reference: Airtable Web API - Rate limits
Jump to: Rate limits (all plans) · Monthly API call limits by plan · Records per request and request size · Records, storage and automation runs by plan · When you hit a limit · Staying under it
Rate limits (all plans)
The two rate limits stack. A single integration writing to one base is capped by the per-base figure, while a tool that fans out across many bases with one person's tokens is capped by the per-token figure. Airtable's recommended response to a 429 is to back off and wait. The official airtable.js client has back-off and retry built in. For high read volume, Airtable suggests a caching proxy instead of more calls.
Monthly API call limits by plan
Monthly call limits are counted per workspace, by calendar month in UTC, and reset on the first day of each month. Every Web API call counts, including metadata and schema calls. Calls made by third-party integrations connected through your account count against the same quota. The two capped plans behave differently at the limit. Free calls are blocked after a one-time 30-day grace period, while Team calls are slowed rather than blocked.
Records per request and request size
Batching is the main throughput lever. Airtable's support documentation says batching handles up to 10 records per request, which at 5 requests per second means up to 50 records per second per base. List requests return pages of at most 100 records, and the whole URL must stay under 16,000 characters. That limit matters when long filterByFormula expressions are URL-encoded. Oversized bodies return 413.
Records, storage and automation runs by plan
These are base and workspace capacity limits rather than API limits, but they are what stops an integration that keeps adding rows. Record counts are per base and include all tables in it. Figures are from Airtable's plans article as of November 2026. Airtable does not publish numeric Enterprise Scale caps there, so they are omitted. Team trials have their own limits, listed on the same page.
| What | Limit | Source |
|---|---|---|
| Records per base | Free 1,000; Team 50,000; Business 125,000 | docs ↗ |
| Attachment storage per base | Free 1 GB; Team 20 GB; Business 100 GB | docs ↗ |
| Automation runs per workspace per month | Free 100; Team 25,000; Business 100,000 | docs ↗ |
| Revision and snapshot history | Free 2 weeks; Team 1 year; Business 1 year | docs ↗ |
| Prices (as of November 2026) | Team $20 per seat/month billed annually ($24 monthly); Business $45 annually ($54 monthly) | docs ↗ |
What happens when you hit a limit
Exceeding the per-base or per-token rate limit returns:
429 Too Many Requests
{
"error": {
"type": "RATE_LIMIT_REACHED",
"message": "Rate limit exceeded. Please try again later"
}
}Requests succeed again after 30 seconds. Airtable advises integrations to back off and wait before retrying.
The monthly call limit works differently. On Free, after the one-time 30-day grace period, calls over 1,000 are blocked until the first of the next month. On Team, calls over 100,000 are not blocked but slow to 2 requests per second until the month resets. An oversized request body returns 413 Request Entity Too Large, and a base at its record limit can't accept new records until you delete some or upgrade.
Staying under the limits
Batch every write. Send 10 records per create, update or delete call; that alone is a 10x reduction in calls. Use upserts (performUpsert) instead of a read followed by a write.
Throttle to 5 per second per base with one shared queue per base, and on 429 pause a full 30 seconds. airtable.js does this for you.
Read less often. Fetch 100 records per page, filter server-side with filterByFormula, request only the fields you need, and cache results. Airtable itself recommends a caching proxy for heavy reads.
Watch the monthly meter. On Free and Team, polling integrations can use up the quota quietly. Lengthen polling intervals or upgrade to Business, which has no monthly cap.
Definitions
- Rate LimitingRate limiting is a technique that controls the number of requests a client can make to a server within a given
- ThroughputThroughput is the amount of work or data processed by a system within a given time period. In automation, it m
- API IntegrationAPI integration is the process of connecting two or more applications through their APIs so they can exchange
Other platforms
Frequently asked questions
What is the Airtable API rate limit?
5 requests per second per base, and 50 requests per second for all traffic using one user's or service account's personal access tokens. Exceeding either returns HTTP 429 with 'Rate limit exceeded. Please try again later', and you must wait 30 seconds before subsequent requests succeed. The limits apply on every plan.
How many API calls does the Airtable Free plan include?
1,000 API calls per workspace per month. After a one-time 30-day grace period, calls over the limit are blocked until the month resets on the first. Every Web API call counts, including schema calls and calls made by third-party tools connected to your account.
What happens when a Team workspace exceeds 100,000 API calls?
Calls are not blocked. Airtable slows them to 2 requests per second until the monthly limit resets on the first day of the next calendar month. Business and Enterprise Scale plans have no monthly call cap and are only subject to the per-base rate limit.
How many records can I create in one Airtable API request?
Up to 10 records per create, update or delete request. Combined with the 5 requests per second per base limit, that gives a documented maximum of about 50 records per second per base. Reading is faster: list records returns up to 100 records per page.
How many records can an Airtable base hold?
As of November 2026: 1,000 records per base on Free, 50,000 on Team and 125,000 on Business, counted across all tables in the base. Attachment storage per base is 1 GB, 20 GB and 100 GB respectively. Enterprise Scale limits are not listed on Airtable's plans page.
Why does my Airtable list request fail with a long formula?
Airtable only accepts requests with a URL shorter than 16,000 characters, and URL-encoded filterByFormula expressions grow quickly. Shorten the formula, filter on fewer fields, or send a POST to the listRecords endpoint so the parameters travel in the request body instead of the URL.
Do Zapier or Make calls count toward my Airtable API limit?
Yes. Airtable's help center says calls from third-party integrations connected through your account consume the same monthly API quota as your own calls, and they share the per-base 5 requests per second rate limit. A frequently polling integration can use up a Free or Team workspace's monthly calls.
Pace the work instead of tripping the limit
Autonoly workflows run on a schedule or a trigger with per-node retries and timeouts, so a job can be spread out and retried rather than run in one burst. Platform limits still apply in full; no tool can lift them.