knowledge base platform limits

Please note that some of these limits do not apply on private instances, and some can be customized.





Min/Max interval for a schedule

Min: 30 seconds interval
Max: 30 day interval


Max. payload size (POST to Blend)

64 KB

Max. incoming triggers

Max. 300/minute per customer account rate limit.
Max. 200 in queue.

See rate limit response headers.
Response code 429 if rate limit exceeded.
Response code 400 if queue is full.


Max. payload size (POST to Blend)

64 KB

Max. incoming events

Max. 20/second rate limit.
Max. 1000 in queue.


Max. number of jobs per month

Depends on license, e.g. 1 million jobs per month.

Total number of queued jobs


Regardless on how jobs were created, e.g. via a Triggered Blend URL or via the SaaS API.

Max. number of concurrent jobs (normal Blends)

1 job per Blend
5 jobs per account (default).

Actual limit per account depends on license.

Maximum number of jobs that run at the same time. If the limit is reached, new jobs will wait to be executed.

Max. number of concurrent jobs (via Webhooks)

10 jobs per Blend

Maximum number of webhook jobs that will run at the same time. If the limit is reached, new jobs will wait to be executed.

Max. job duration

See table below

Depends on execution context

Max. memory consumption of a Blend
(normal Blends and realtime Blends)

256 MB

Blends use iterators to process large amounts of data (e.g. millions of records), the memory limit is not reached in normal circumstances.

Failed job alert emails

Max. 5/day per Blend

Sent to Admin users of the account.

API call timeout

60 seconds

Note that uses paging, this applies to a single outbound API call.

Call URL block timeout

User configurable (e.g. 600 seconds)

Max. timeout is defined by max. Blend execution time.

Custom code block max. execution time

Normal Blends: 3600 seconds
Real-time Blends: 60 seconds

Custom code blocks have an overhead of 500 msec per execution when used in real-time Blends.


Max. number of job logs

5000 per Blend
Max. 500MB per Blend

Oldest jobs will be deleted if limits are exceeded.

Job log retention

90 days

Customisable on private instances.

Realtime Blends

No job logs

Realtime Blends do not have job logs.

Max. size of one job log


Job log will be truncated if larger

Max. size of the state of one Block in a job log

1MB per block execution
5MB per block in total (across all executions of the block in the job)

Max. 200 items are stored for a list.

State will be truncated if larger


Maximum API calls

300 calls/minute rate limit per account

This includes API's to the SaaS API, executing Triggered Blends via their API URL etc.


Custom code

Supported languages: PHP, JS, Python

Package/library install or import not supported. See Custom scripts for alternative solutions.

Variable blocks

Variables are stored in memory (see Blend memory limit)

Use Data Store or CDP to persist data or to process large amounts of data.

List blocks (Filter list, compare list)

Full list loaded in memory (see Blend memory limit)

Use loops to process large lists, loops use iterators and avoid loading all items in memory at once.


Maximum records

400K records per account
(sum of all tables)

Data Store:
400K records per account
(sum of all Data Stores)

Maximum Blend job duration (execution time)

Blend execution context

Execution environment

Max. execution time

Manual (Run button in editor & Test runs)


7 days

Run button on Blend detail page (outside of Blend editor)


7 days

Triggered Blend (via API URL)

Depends on Blend setting
"Realtime Blend":

Not realtime: Docker
Realtime: Lambda

Non-realtime (Docker):
Async=false: 1 minute
Async=true: 7 days

Realtime (Lambda):
Async=false: 1 minute
Async=true: 15 minutes


Same as above

Same as above



15 minutes



7 days

Settings, Setup, Delete flows in Templates


7 days

Updated 3 months ago platform limits

Suggested Edits are limited on API Reference Pages

You can only suggest edits to Markdown body content, but not to the API spec.