Hi ABBYY Team / Community,
I’m working on a solution using New Document AI API and I wanted to clarify a key capability:
Can the new Document AI API automatically classify incoming documents by type (e.g., Invoice, Utility Bill, Bank Statement, ID, Tax Form, etc.) before extraction?
Specifically, I’m looking for information on:
-
Whether Document AI API provides pre-trained classification skills or supports building custom classification models?
-
Can this classification happen before data extraction?
-
-
Any best practices on handling multi-document PDFs with mixed document types?
Would appreciate any guidance, documentation, or examples on how to implement this.
Comments
1 comment
Hi Vivek! Great questions. Unfortunately at the moment the API can't classify documents but that's on our radar. Given what you're looking to do (such as mixed document types), I suggest looking into ABBYY Vantage which has classification and much more. If you need access for the upcoming ABBYY DevCon India 2025, I can grant access to a trial instance during the hackathon period. just let me know! cheers
Please sign in to leave a comment.