Rate limits

By default, each account can send up to 6,000 search queries or fetched URLs a minute. Telem has no daily or monthly limit and no spending cap. Each request also has size limits.

Per-minute limit

By default, each account can send 6,000 items a minute, shared by all its keys. An item is:

• one query of a search. A batch of 10 queries counts 10.
• one URL of a fetch. A fetch of 5 URLs counts 5.
• one POST to /v1/monitors or below it, such as creating or polling a monitor.

Other calls, such as reading a session or listing your monitors, don’t count.

The count starts again at the start of each minute. A call that would take the account over the limit gets a 429:

{"detail": "Rate limit exceeded (6000 requests/minute for this account)"}

Retry-After gives the seconds until the next minute, from 1 to 60. A call that Telem turns away still counts toward the minute, so wait for the next minute before you send it again.

No daily or monthly limit

Telem does not cap how many calls you make in a day or a month, and it does not cap how much you spend. Your prepaid credit pays for the calls. See Pricing.

Limits on a key

A key can also have its own requests-per-minute cap, but only if one was set for your project or for the key. This cap counts every call made with the key, one for each call. Over it, a call gets a 429 with Retry-After: 60:

{"detail": "Rate limit exceeded (60 requests/minute for this key)"}

If you get a 429

The message names the limit you reached: for this account or for this key. Wait for the number of seconds in Retry-After, then send the call again.

• The JavaScript SDK does this for you: it retries a 429 twice by default (maxRetries), waiting as Retry-After says, up to 30 seconds each time. Then it throws TelemError.
• The Python SDK raises APIStatusError with status_code 429. Wait and retry in your own code.

To stay under the limit, spread large jobs over several minutes, and limit how many calls run at once. See Search best practices.

Request limits

These limits apply to each request. Telem checks them before it bills the request, so a rejected request costs nothing. It still counts toward the per-minute limit.

Limit	Value	Applies to
URLs per fetch	5	one POST /v1/fetch call. One more is a 400 (over_url_cap)
Results per provider	1 to 20	num_results
Queries per batch	32	a list of queries
Volume budget	160	queries × results, on every search

Next steps

• Pricing: what each call costs
• Errors: every status code and what to do
• Feedback: ask a question about limits