These pages document Vantage Cloud. To run Vantage on your own Kubernetes infrastructure instead, see ABBYY Vantage 3.0 self-hosted.
Process via API
Authenticate, upload a document, and extract data using the REST API. Best for developers integrating Vantage into applications. ~10 minutes.
Process via UI
Upload a document and review extracted data in the Vantage web interface. Best for business users and evaluation. ~10 minutes.
How Vantage processes a document
Whether you use the API or the web UI, every document goes through the same pipeline:1
Authenticate or sign in
Connect to your Vantage tenant using API credentials or the web login.
2
Select a skill
A skill is an AI model trained to extract data from a specific document type. Vantage has 100+ pre-trained skills, or you can build your own.
3
Submit your document
Upload a PDF, image, or Office document. Vantage creates a transaction, a processing job that tracks the document through extraction.
4
Extract structured data
The skill reads the document and extracts field values (e.g., invoice number, vendor name, total amount). Each field gets a confidence score from 0 to 100 indicating how certain Vantage is about the extraction.
5
Review and export results
Download the extracted data as JSON or XML, or verify it in Manual Review, a built-in interface where operators can check and correct fields before the data moves downstream.
What you’ll need
- An ABBYY Vantage account (contact ABBYY to request a tenant)
- A sample document (invoice, receipt, purchase order, or any supported type)
- About 10 minutes