Parabola uses a credit-based system to track overall usage in the platform. Credits are consumed when processing rows of data in Parabola flows. Understanding how credits work can help you optimize your flows and manage your usage effectively.
Credits Overview
-
Credits are used each time you run a flow or perform AI-driven tasks.
-
Credits are not used when building or editing a flow
-
Credits are only used by steps that have non-zero input/output rows. For example, filtering to zero rows or an error in your flow stops credit consumption
-
Credit usage is calculated by:
-
Steps without AI: 10 credits per step, regardless of row count.
- Exceptions here are the “Send Email by Rows” and “Fill Addresses” which use additional credits on the basis of row count
-
AI steps: 10 credits plus variable credits based on prompt complexity and data size
-
PDF parsing: credits vary by logic complexity.
When Credits Are Consumed
-
Credits are only consumed by steps that successfully process rows of data
-
Disconnected or error-producing steps do not consume credits
-
Each processed row counts towards credit consumption
Monitoring Credit Usage
You can track your credit consumption by visiting the Billing page. This page shows your current credit balance and usage history.
The billing page only shows total credit usage across your team for the entire billing period. This counter on the billing page updates in real-time, so if you record the number of credits before and after running a flow, you should be able to estimate the number of credits it took to run the flow.
Currently, there is no way to view the number of credits a single user, flow, or flow run consumes for a given billing period. For analytical purposes, the Usage Reporting & Analytics dashboard can provide additional insights into credit usage.
Optimizing Credit Usage
To minimize credit consumption in your flows:
-
Filter out unnecessary rows early in your flow
-
Remove or disconnect steps that aren’t needed for your final output
-
Consider using filtering or limiting steps at the beginning of your flow to reduce the number of rows being processed
-
Be mindful of frequency, unnecessary or repetitive flow runs can result in excess credit usage
The earlier you filter out unnecessary data in your flow, the fewer credits you’ll consume in some subsequent steps.
