ID reader prebuilt model

You can use the identity document (ID) reader prebuilt model to extract information from passports and US driver licenses, social security, and green cards. The model extracts information such as the person’s first name, date of birth, or gender.

Images such as scans or pictures of the identity documents are deleted once processed by the model.

Note

  • The design and format of identity documents vary widely. AI Builder is constantly improving the accuracy of the ID reader prebuilt model, but it's possible there could be inaccurate or missing information in some cases. It's a good idea to verify that the output is as you expect.
  • The ID reader prebuilt model only supports Latin character extraction at this time.

Licensing requirements

AI Builder is licensed as an add-on to your Power Apps or Power Automate licenses. For information about license capacity, pricing, and restrictions, see AI Builder licensing.

Role requirements

Users need to have the Basic User role to consume the ID reader prebuilt model.

Use in Power Automate

If you want to use this model in Power Automate, you can find more information in Use the ID reader prebuilt model in Power Automate.

Supported language, format, and size

The images you can process with the ID reader prebuilt model must have these characteristics:

  • Format:
    • jpg
    • png
    • pdf
  • Size: 20 MB maximum (use small images for fastest processing time)
  • For PDF documents, only the first 2,000 pages are processed.

Model output

If a valid identity document is detected, the model tries to locate and extract the following properties.

Property Note
First name N/A
Last name N/A
Gender N/A
Date of birth N/A
Place of birth US green card only
Region US driver license only
Country N/A
Nationality Passport only
Street address US driver license only
Identity document number N/A
Date of issue N/A
Identity document expiration date N/A
Category code US green card only

Limits

Action Limit Renewal period
Identity document reader calls (per environment) 24 60 seconds