ML models and capabilities

All of our models can be trained to understand any language recognized by an OCR. This also applies to languages not immediately recognized by the model. For instance, if you're working with purchase orders in Romanian, you'll need to specifically train the model with Romanian language purchase orders. Check the OCR page for a list of languages supported by our OCR engines.

Important: Exceptions are Chinese, Japanese, Korean and right-to-left languages. To process documents in these languages, you can only train them on the out-of-the-box model package specifically designed for that language (like Invoices China, Invoices Hebrew, or Invoices Japan). Alternatively, you can use the generic Document Understanding package.

The following table lists the languages supported out-of-the-box by our models.

ML model or capability	Languages supported out of the box (pre-trained)
Form Extractor	The same languages supported by the OCR engine
Intelligent Keyword Classifier	The same languages supported by the OCR engine
Machine Learning Classifier	The same languages supported by the model used
Machine Learning Extractor	The same languages supported by the model used
709 - Preview	English
941x - Preview	English
1040	English
1040 Schedule C - Preview	English
1040 Schedule D - Preview	English
1040 Schedule E - Preview	English
1040x - Preview	English
3949a - Preview	English
9465 - Preview	English
4506T	English
Acord125	English
Acord126	English
Acord131	English
Acord140	English
Acord25	English
Bank Statements	English
Bills of Lading	English German
Certificate of Incorporation/ Good Standing	English
Certificate of Origin	English
Checks	English
Children Product Certificate	English
CMS1500	English
EU Declaration of Conformity	English
Financial Statements	English
FM1003 - Preview	English
I9	English
IDCards	Trained to support ID cards and driver's licenses from: Aadhaar Australia Austria Belgium Canada Croatia Cyprus Finland France Germany Hong Kong Hungary India Italy Netherlands PAN cards Poland Romania Saudi Arabian Spain Switzerland United Kingdom United States of America (all 50 states, including Washington D.C.)
Invoices	English French German Portuguese Romanian Spanish
Note: Deprecated: model has been merged into the Invoices model. See the Invoices model page for details.	English
Invoices China	Chinese - Simplified Chinese - Traditional
Invoices Hebrew - Preview	Hebrew
Invoices India	English
Invoices Japan	Japanese
Invoices Shipping	English
Packing Lists	English
Passports	All nationalities
Pay slips	English
Purchase Orders	English German
Receipts	English Finnish French German Norwegian Romanian Spanish
RemittanceAdvices	English
UB04 - ML package - Preview	English
Utility Bills	English
Vehicle Titles	English
W2	English
W9	English Spanish