Custom Classifier - AWS Comprehend - Alternative Options

Question

We were looking to use AWS Comprehend custom classifier but its pricing seems way high as it starts charging the moment is put and even if not used ("Endpoints are billed on one second increments, with a minimum of 60 seconds. Charges will continue to incur from the time you start the endpoint until it is deleted even if no documents are analyzed.")

So, we need the feature but would like to see if there is an alternate way to use the classifiers we have.

Any ideas?

Answer 1

Comprehend supports both synchronous and asynchronous inference on custom classifiers. Synchronous inference provides sub-second response time but requires setting up a custom endpoint to host the model and is charged on uptime.

Asynchronous inference (StartDocumentClassificationJob) usually takes a few minutes to an hour dependent on the amount of data being processed and is billed based on data volume (1 billing units = 100 characters).