Processor list

This page contains detailed information on all processors offered by Document AI. You can see a list of all processors by solution type.

All Document AI processors adhere to the Data Processing and Security Terms.

Custom processors

Refer to Sample datasets for sample labeled and unlabeled datasets to use for training.

Custom Extractor

Description

Extract fields from documents using generative AI or custom models; fine tune models to accurately extract data from your documents.

Category Custom
Functions OCR, Entity Extraction
Release stage General availability
Access status Public
Type in API CUSTOM_EXTRACTION_PROCESSOR
Notes
  • If using generative AI for extraction, then:

    • Only the English language is officially supported.
    • Region availability is in the US, EU, northamerica-northeast1 and asia-southeast1.

Supported languages
Full list of languages
Language Name BCP 47 Tag Script Handwriting supported
Afrikaans af Latn
Arabic ar Arab
Azerbaijani az Latn
Azerbaijani (Cyrillic) az-Cyrl Cyrl
Belarusian be Cyrl
Bulgarian bg Cyrl
Bosnian bs Latn
Catalan ca Latn
Cebuano ceb Latn
Czech cs Latn
Welsh cy Latn
Danish da Latn
German de Latn
Greek el Grek
English en Latn
Esperanto eo Latn
Spanish es Latn
Estonian et Latn
Basque eu Latn
Persian fa Arab
Finnish fi Latn
Filipino fil Latn
French fr Latn
Irish ga Latn
Galician gl Latn
Hindi hi Deva
Croatian hr Latn
Haitian Creole ht Latn
Hungarian hu Latn
Indonesian id Latn
Icelandic is Latn
Italian it Latn
Hebrew iw Hebr
Japanese ja Jpan
Javanese jv Latn
Kazakh kk Cyrl
Korean ko Kore
Kyrgyz ky Cyrl
Latin la Latn
Lithuanian lt Latn
Latvian lv Latn
Macedonian mk Cyrl
Mongolian mn Cyrl
Marathi mr Deva
Malay ms Latn
Maltese mt Latn
Nepali ne Deva
Dutch nl Latn
Norwegian no Latn
Polish pl Latn
Pashto ps Arab
Portuguese (Portugal & Brazil) pt Latn
Romanian ro Latn
Russian ru Cyrl
Russian (Petrine Orthography) ru-PETR1708 Cyrl
Sanskrit sa Deva
Slovak sk Latn
Slovenian sl Latn
Albanian sq Latn
Serbian sr Cyrl
Swedish sv Latn
Swahili sw Latn
Tagalog tl Latn
Turkish tr Latn
Ukrainian uk Cyrl
Urdu ur Arab
Uzbek uz Latn
Uzbek (Cyrillic) uz-Cyrl Cyrl
Vietnamese vi Latn
Yiddish yi Hebr
Chinese simplified zh-Hans Hani
Chinese traditional zh-Hant Hani
Zulu zu Latn
Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 200
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-south1
  • asia-southeast1
  • australia-southeast1
  • eu
  • europe-west2
  • europe-west3
  • northamerica-northeast1
  • us

Custom Classifier

Description

Train a model to classify a document type from a set of classes.

Category Custom
Functions OCR, Classification
Release stage General availability
Access status Public
Type in API CUSTOM_CLASSIFICATION_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 200
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-south1
  • asia-southeast1
  • australia-southeast1
  • eu
  • europe-west2
  • europe-west3
  • northamerica-northeast1
  • us
More information Create a custom classification processor

Custom Splitter

Description

Train a model to split a file containing multiple documents into individual, classified documents.

Category Custom
Functions OCR, Classification, Splitting
Release stage General availability
Access status Public
Type in API CUSTOM_SPLITTING_PROCESSOR
Notes
  • i18n can be supported via custom training options, not through UI.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 1000
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-south1
  • asia-southeast1
  • australia-southeast1
  • eu
  • europe-west2
  • europe-west3
  • northamerica-northeast1
  • us
More information Create a custom splitter processor

Summarizer

Description

Get abstract and bullet point summaries for short and long documents

Category Custom
Functions Summarize
Release stage Preview
Access status Public
Type in API SUMMARY_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-foundation-model-v1.0-2023-08-22 Stable

None

None

Google foundation model

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 250
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • us

General processors

Enterprise Document OCR (Optical Character Recognition)

Description

Identify and extract text in different types of documents.

This processor allows you to identify and extract text, including handwritten text, from documents in over 200 languages. The processor also uses machine learning to perform a quality assessment of a document based on the readability of its content.

Category General
Functions OCR, Quality Analysis
Release stage General availability
Access status Public
Type in API OCR_PROCESSOR
Supported languages
Full list of languages
Language Name BCP 47 Tag Script Handwriting supported
Afrikaans af Latn
Albanian sq Latn
Arabic ar Arab
Armenian hy Armn
Belarusian be Cyrl
Bangla bn Beng
Bengali bn Beng
Bulgarian bg Cyrl
Catalan ca Latn
Chinese zh Hani
Croatian hr Latn
Czech cs Latn
Danish da Latn
Dutch nl Latn
English en Latn
Estonian et Latn
Filipino fil Latn
Finnish fi Latn
French fr Latn
German de Latn
Greek el Grek
Gujarati gu Gujr
Hebrew iw Hebr
Hindi hi Deva
Hungarian hu Latn
Icelandic is Latn
Indonesian id Latn
Italian it Latn
Japanese ja Jpan
Kannada kn Knda
Khmer km Khmr
Korean ko Kore
Lao lo Laoo
Latvian lv Latn
Lithuanian lt Latn
Macedonian mk Cyrl
Malay ms Latn
Malayalam ml Mlym
Marathi mr Deva
Nepali ne Deva
Norwegian no Latn
Persian fa Arab
Polish pl Latn
Portuguese (Portugal & Brazil) pt Latn
Punjabi pa Guru
Romanian ro Latn
Russian ru Cyrl
Serbian sr Cyrl
Slovak sk Latn
Slovenian sl Latn
Spanish es Latn
Swedish sv Latn
Tagalog tl Latn
Tamil ta Taml
Telugu te Telu
Thai th Thai
Turkish tr Latn
Ukrainian uk Cyrl
Vietnamese vi Latn
Yiddish yi Hebr
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-ocr-v2.0-2023-06-02 Stable

None

None

Production-ready model specialized for document use cases. Includes access to all OCR add-ons.
pretrained-ocr-v1.2-2022-11-10 Stable

None

None

Frozen model version of v1.0: Model files, configurations, and binaries of a version snapshot frozen in a container image for up to 18 months.
pretrained-ocr-v1.1-2022-09-12 Stable

None

None

Not recommended for use and will be deprecated.
pretrained-ocr-v1.0-2020-09-23 Stable

None

None

Not recommended for use and will be deprecated.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 500
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-south1
  • asia-southeast1
  • australia-southeast1
  • eu
  • europe-west2
  • europe-west3
  • northamerica-northeast1
  • us
More information Enterprise Document OCR

Form Parser

Description

Extract general key-value pairs (entity and checkbox), tables, and generic entities from documents in addition to OCR text.

This processor applies advanced machine learning technologies to extract key-value pairs, checkboxes and tables from documents in over 200+ languages. This processor also leverages deep learning models to extract 11 generic entities that are common in various document types.

Category General
Functions OCR, Form Parsing, Entity Extraction
Release stage General availability
Access status Public
Type in API FORM_PARSER_PROCESSOR
Supported languages
Full list of languages
Language Name BCP 47 Tag Script Handwriting supported
Afrikaans af Latn
Albanian sq Latn
Arabic ar Arab
Belarusian be Cyrl
Catalan ca Latn
Chinese zh Hani
Croatian hr Latn
Czech cs Latn
Danish da Latn
Dutch nl Latn
English en Latn
Estonian et Latn
Filipino fil Latn
Finnish fi Latn
French fr Latn
German de Latn
Hebrew iw Hebr
Hindi hi Deva
Hungarian hu Latn
Icelandic is Latn
Indonesian id Latn
Italian it Latn
Japanese ja Jpan
Korean ko Kore
Latvian lv Latn
Lithuanian lt Latn
Macedonian mk Cyrl
Malay ms Latn
Marathi mr Deva
Nepali ne Deva
Norwegian no Latn
Persian fa Arab
Polish pl Latn
Portuguese (Portugal & Brazil) pt Latn
Romanian ro Latn
Russian ru Cyrl
Serbian sr Cyrl
Slovak sk Latn
Slovenian sl Latn
Spanish es Latn
Swedish sv Latn
Tagalog tl Latn
Turkish tr Latn
Ukrainian uk Cyrl
Vietnamese vi Latn
Yiddish yi Hebr
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-form-parser-v1.0-2020-09-23 Stable

None

None

Legacy version. For best quality and full feature set, use the Form Parser v2.0.
pretrained-form-parser-v2.0-2022-11-10 Stable
Show fields
  • email
  • phone
  • url
  • date_time
  • address
  • person
  • organization
  • quantity
  • price
  • id
  • page_number

None

Recommended version. Supports generic entities and includes upgraded table, KVP, and checkbox model, as well as more than 200 languages.
pretrained-form-parser-v2.1-2023-06-26 Release Candidate

None

None

Public Preview version. Same model as v2.0 with native text extraction from digital PDF files enabled.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 100
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-south1
  • asia-southeast1
  • australia-southeast1
  • eu
  • europe-west2
  • europe-west3
  • northamerica-northeast1
  • us
More information Form Parser

Intelligent Document Quality Processor

Description

Perform quality assessment of a document based on its readability and get a quality score.

Intelligent Document Quality processor uses machine learning to perform quality assessment of a document based on the readability of its content. This quality assessment is returned as a quality score [0, 1], where 1 means perfect quality. If the quality score detected is lower than 0.5, a list of negative quality reasons (sorted by the likelihood) is also returned.

Category General
Functions OCR, Quality Analysis
Release stage Preview
Access status Limited [3]
Type in API DOCUMENT_QUALITY_PROCESSOR
Notes
  • Quality score is returned in the confidence field of the entity with type="quality_score".
  • The quality/defect_* properties are sorted in descending order by confidence value.
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-document-quality-v1.0-2021-01-20 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 5
Maximum pages (batch/offline/asynchronous requests): 100
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • quality_score
  • quality/defect_blurry
  • quality/defect_dark
  • quality/defect_faint
  • quality/defect_noisy
  • quality/defect_text_too_small
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

Document Splitter

Description

Programmatically split documents on logical boundaries.

Document Splitter uses machine learning to separate documents on logical boundaries. For example, if you have one PDF document with multiple scanned files, the Document AI API will suggest the page location of a new file.

Category General
Functions OCR, Splitting
Release stage Preview
Access status Limited [3]
Type in API DOCUMENT_SPLIT_PROCESSOR
Notes
  • Maximum image size: 65500 x 65500 pixels
  • The splitter is not designed to split logical documents that are over 30 pages long. Logical documents that are more than 30 pages long (e.g. a 40-page bank statement) may be split into two or more docs.
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-document-split-v1.0-2020-09-20 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 2000
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us
More information Document splitters behavior

Specialized processors

France Driver License Parser

Description

Extract fields such as names, document ID, date of birth, etc.

Category Specialized
Solution type Identity
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FR_DRIVER_LICENSE_PROCESSOR
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-fr-driver-license-v1.0-2021-06-14 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 2
Maximum pages (batch/offline/asynchronous requests): 2
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • Family Name
  • Given Names
  • Document Id
  • Expiration Date
  • Date Of Birth
  • Issue Date
  • Portrait
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

France National ID Parser

Description

Extract fields such as names, document ID, date of birth, etc.

Category Specialized
Solution type Identity
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FR_NATIONAL_ID_PROCESSOR
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-fr-national-id-v1.0-2021-06-14 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 2
Maximum pages (batch/offline/asynchronous requests): 2
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • Family Name
  • Given Names
  • Document Id
  • Expiration Date
  • Date Of Birth
  • Issue Date
  • Address
  • Portrait
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • Date Of Birth
  • Expiration Date
  • Issue Date
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

France Passport Parser

Description

Extract fields such as names, document ID, date of birth, etc.

Category Specialized
Solution type Identity
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FR_PASSPORT_PROCESSOR
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-fr-passport-v1.0-2022-04-29 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 2
Maximum pages (batch/offline/asynchronous requests): 2
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • family_name
  • given_name
  • document_id
  • expiration_date
  • date_of_birth
  • issue_date
  • address
  • place_of_birth
  • portrait
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • Date Of Birth
  • Expiration Date
  • Issue Date
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

Identity Document Proofing Parser

Description

Predict the validity of ID documents using multiple signals.

Identity Document Proofing Processor is designed to help predict the validity of ID documents with four different signals.

The processor currently returns information from the following signals:

  • is_identity_document detection: Predicts whether an image contains a recognized identity document.
  • suspicious_words detection: Predicts whether words are present that aren't typical on IDs.
  • image_manipulation detection: Predicts whether the image was altered or tampered via an image editing tool.
  • online_duplicate detection: Predicts whether the image can be found online.

Category Specialized
Solution type Identity
Functions OCR, Quality Analysis
Release stage Preview
Access status Public
Type in API ID_PROOFING_PROCESSOR
Notes
  • The Online Duplicate Detection feature is currently processed in US data centers. Regional and multi-regional support is unavailable for this feature outside of the US.
  • This processor is supported by algorithms that are updated more frequently than new processor versions are released. For this reason, the processor might return different outputs over time even when using the same processor version. For example, the Online Duplicate Detection system monitors images present on the web. The system's behavior can then change more quickly than can be tracked in processor versions.
  • Refer to notes on Responsible AI [4] and Human review [5]
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • Support for US passports, passcards and driver's licenses.
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-id-proofing-v1.0-2022-10-03 Stable

None

None

pretrained-id-proofing-v1.1-2023-05-18 Release Candidate
Show fields
  • fraud_signals_photocopy_detection

None

Additional photocopy detection signal

pretrained-id-proofing-v1.2-2023-10-04 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 2
Maximum pages (batch/offline/asynchronous requests): 2
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • fraud_signals_is_identity_document
  • fraud_signals_suspicious_words
  • evidence_suspicious_word
  • evidence_inconclusive_suspicious_word
  • fraud_signals_image_manipulation
  • fraud_signals_online_duplicate
  • evidence_hostname
  • evidence_thumbnail_url
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • fraud_signals_image_manipulation
  • fraud_signals_online_duplicate
  • fraud_signals_is_identity_document
  • fraud_signals_suspicious_words
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

US Driver License Parser

Description

Extract fields such as names, document ID, date of birth, etc.

Category Specialized
Solution type Identity
Functions OCR, Entity Extraction
Release stage General availability
Access status Public
Type in API US_DRIVER_LICENSE_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • Supports all 50 States and D.C.
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-us-driver-license-v1.0-2021-06-14 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 2
Maximum pages (batch/offline/asynchronous requests): 2
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • Family Name
  • Given Names
  • Document Id
  • Expiration Date
  • Date Of Birth
  • Issue Date
  • Address
  • Portrait
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • Date Of Birth
  • Expiration Date
  • Issue Date
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

US Passport Parser

Description

Extract fields such as names, document ID, date of birth, etc.

Category Specialized
Solution type Identity
Functions OCR, Entity Extraction
Release stage General availability
Access status Public
Type in API US_PASSPORT_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-us-passport-v1.0-2021-06-14 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 2
Maximum pages (batch/offline/asynchronous requests): 2
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • Family Name
  • Given Names
  • Document Id
  • Expiration Date
  • Date Of Birth
  • Issue Date
  • MRZ Code
  • Portrait
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • Date Of Birth
  • Expiration Date
  • Issue Date
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

1003 Parser

Description

Extract over 50 fields from Fannie Mae Form 1003 (URLA).

The 1003 Form is Fannie Mae's form number for the Uniform Residential Loan Application (URLA), a borrower’s application for a mortgage. Freddie Mac's form number is Form 65; both refer to the same form.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1003_PROCESSOR
Notes
  • Batch processing currently not available for this processor.
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • Legacy Form 1003 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1003-v1.0-2020-10-01 Stable

None

None

pretrained-1003-v1.1-2021-12-10 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AmortizationType
  • BuiltYear
  • HasPrimaryBorrowerSignaturePage1
  • HasSecondaryBorrowerSignaturePage1
  • LoanAmount
  • LoanPurpose
  • LoanTerm
  • MortgageType
  • NoteRate
  • PrimaryBorrowerCurrentJobTitleYearsTerm
  • PrimaryBorrowerDateOfBirth
  • PrimaryBorrowerDependentAge
  • PrimaryBorrowerDependentCount
  • PrimaryBorrowerDependentCountAndAge
  • PrimaryBorrowerEmployerNameAndAddress
  • PrimaryBorrowerHomePhone
  • PrimaryBorrowerJobTitle
  • PrimaryBorrowerJobYearsTerm
  • PrimaryBorrowerMarrageStatus
  • PrimaryBorrowerName
  • PrimaryBorrowerSSN
  • PrimaryBorrowerSelfEmployedIndicator
  • PrimaryBorrowerWorkPhoneNumber
  • PropertyAddress
  • PropertyUsageType
  • RefinanceLoanInfo
  • RefinancePurposeType
  • SecondaryBorrowerCurrentJobTitleYearsTerm
  • SecondaryBorrowerDateOfBirth
  • SecondaryBorrowerDependentAge
  • SecondaryBorrowerDependentCount
  • SecondaryBorrowerDependentCountAndAge
  • SecondaryBorrowerEmployerNameAndAddress
  • SecondaryBorrowerHomePhone
  • SecondaryBorrowerJobTitle
  • SecondaryBorrowerJobYearsTerm
  • SecondaryBorrowerMarrageStatus
  • SecondaryBorrowerName
  • SecondaryBorrowerSSN
  • SecondaryBorrowerSelfEmployedIndicator
  • SecondaryBorrowerWorkPhoneNumber
  • Units
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

1040 Parser

Description

Extract from Form 1040, including name, filing status, amounts, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1040_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2021 (pretrained-1040-v2.0-2022-08-24 version only)
  • 2020 (pretrained-1040-v2.0-2022-08-24 version only)
  • 2019
  • 2018
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1040-v1.0-2020-10-01 Stable

None

None

pretrained-1040-v1.1-2021-12-10 Stable

None

None

pretrained-1040-v2.0-2022-08-24 Release Candidate
Show fields
  • pensions_annuities_taxable
  • social_security_benefits_taxable
  • ira_distributions
  • ira_distributions_taxable_amount
  • qualified_dividends
  • ordinary_dividends
  • tax_exempt_interest
  • taxable_interest

None

Quality improvements.

Added support for Year 2020 and 2021.

Breaking change: the names of all extracted fields have been renamed from CamelCase to snake_case (for example, first_name instead of FirstName).

This change was made to standardize the format of field names across Document AI.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AddressApt
  • AddressCityStateZip
  • AddressStreet
  • CapitalGain
  • EmailAddress
  • FilingStatusCheckbox
  • FirstName
  • LastName
  • Occupation
  • OtherIncome
  • PensionsAnnuities
  • PhoneNumber
  • SSN
  • SocialSecurityBenefits
  • SpouseFirstName
  • SpouseLastName
  • SpouseOccupation
  • SpouseSSN
  • TotalIncome
  • WagesSalariesTips
  • Year
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

1040 Schedule C Parser

Description

Extract from Form 1040 Schedule C, including name, wages, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1040SCH_C_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1040sch-c-v1.0-2021-05-27 Stable

None

None

pretrained-1040sch-c-v1.1-2021-12-10 Stable

None

None

pretrained-1040sch-c-v2.0-2022-11-07 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • ProprietorName
  • EmployerIdentificationNumber
  • Wages
  • SocialSecurityNumber
  • BusinessName
  • BusinessAddress
  • OtherIncome
  • Depletion
  • DepreciationAndSection179ExpenseDeduction
  • DeductibleMeals
  • ExpensesForBusinessUseOfYourHome
  • NetProfitOrLoss
  • TotalMilesDrivenWithVehicleForBusiness
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

1040 Schedule D Parser

Description

Extract from Form 1040 Schedule D, including name, gains, losses, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_1040SCH_D_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1040sch-d-v1.0-2021-11-17 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • amount_if_line_16_is_loss
  • amount_unrecaptured_section_1250_gain
  • are_lines_15_and_16_both_gains_checkbox
  • are_lines_18_and_19_both_zero_or_blank_checkbox
  • capital_gain_distributions
  • combine_net_short_and_long_term_capital_gain_loss
  • dispose_of_investment_during_tax_year_checkbox
  • long_term_capital_loss_carryover
  • long_term_gain_or_loss
  • name
  • net_long_term_capital_gain_or_loss
  • net_long_term_gain_or_loss
  • net_short_term_capital_gain_or_loss
  • net_short_term_gain_or_loss
  • qualified_dividends_checkbox
  • short_term_capital_loss_carryover
  • short_term_gain
  • social_security_number
  • totals_long_term_transactions_boxd_adjustments
  • totals_long_term_transactions_boxd_cost
  • totals_long_term_transactions_boxd_gain_or_loss
  • totals_long_term_transactions_boxd_proceeds
  • totals_long_term_transactions_boxe_adjustments
  • totals_long_term_transactions_boxe_cost
  • totals_long_term_transactions_boxe_gain_or_loss
  • totals_long_term_transactions_boxe_proceeds
  • totals_long_term_transactions_boxf_adjustments
  • totals_long_term_transactions_boxf_cost
  • totals_long_term_transactions_boxf_gain_or_loss
  • totals_long_term_transactions_boxf_proceeds
  • totals_long_term_transactions_cost
  • totals_long_term_transactions_gain_or_loss
  • totals_long_term_transactions_proceeds
  • totals_short_term_transactions_boxa_adjustments
  • totals_short_term_transactions_boxa_cost
  • totals_short_term_transactions_boxa_gain_or_loss
  • totals_short_term_transactions_boxa_proceeds
  • totals_short_term_transactions_boxb_adjustments
  • totals_short_term_transactions_boxb_cost
  • totals_short_term_transactions_boxb_gain_or_loss
  • totals_short_term_transactions_boxb_proceeds
  • totals_short_term_transactions_boxc_adjustments
  • totals_short_term_transactions_boxc_cost
  • totals_short_term_transactions_boxc_gain_or_loss
  • totals_short_term_transactions_boxc_proceeds
  • totals_short_term_transactions_cost
  • totals_short_term_transactions_gain_or_loss
  • totals_short_term_transactions_proceeds
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

1040 Schedule E Parser

Description

Extract from Form 1040 Schedule E, including name, expenses, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1040SCH_E_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1040sch-e-v1.0-2021-04-14 Stable

None

None

pretrained-1040sch-e-v1.1-2021-12-10 Stable

None

None

pretrained-1040sch-e-v2.0-2022-05-05 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • NamesShownOnReturn
  • YourSocialSecurityNumber
  • PhysicalAddressOfEachProperty_A
  • PhysicalAddressOfEachProperty_B
  • PhysicalAddressOfEachProperty_C
  • TypeOfProperty_A
  • TypeOfProperty_B
  • TypeOfProperty_C
  • FairRentalDays_A
  • FairRentalDays_B
  • FairRentalDays_C
  • PersonalUseDays_A
  • PersonalUseDays_B
  • PersonalUseDays_C
  • RentsReceived_A
  • RentsReceived_B
  • RentsReceived_C
  • RoyaltiesReceived_A
  • RoyaltiesReceived_B
  • RoyaltiesReceived_C
  • Insurance_A
  • Insurance_B
  • Insurance_C
  • MortgageInterest_A
  • MortgageInterest_B
  • MortgageInterest_C
  • Taxes_A
  • Taxes_B
  • Taxes_C
  • Depletion_A
  • Depletion_B
  • Depletion_C
  • OtherExpenses_A
  • OtherExpenses_B
  • OtherExpenses_C
  • TotalExpenses_A
  • TotalExpenses_B
  • TotalExpenses_C
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

1099-DIV Parser

Description

Extract from Form 1099-DIV, including account number, qualified dividends, federal income tax withheld, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1099DIV_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
  • 2018 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1099div-v1.0-2021-05-27 Stable

None

None

pretrained-1099div-v1.1-2021-12-10 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • PayerTIN
  • RecipientTIN
  • RecipientName
  • RecipientAddressLine1
  • RecipientAddressLine2
  • PayerName
  • PayerAddressAndPhone
  • AccountNumber
  • TotalOrdinaryDividends
  • QualifiedDividends
  • TotalCapitalGainDistribution
  • UnrecapSec1250Gain
  • Section1202Gain
  • CollectiblesGain
  • NondividendDistributions
  • FederalIncomeTaxWithheld
  • Section199ADividends
  • CashLiquidationDistributions
  • NoncashLiquidationDistributions
  • ExemptInterestDividends
  • SpecifiedPrivateActivityBondInterestDividends
  • FatcaFilingRequirement
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

1099-G Parser

Description

Extract from Form 1099-G, including payer, recipient, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1099G_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1099g-v1.0-2021-05-27 Stable

None

None

pretrained-1099g-v1.1-2021-12-10 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • PayerName
  • PayerAddressAndPhone
  • PayerTIN
  • RecipientTIN
  • RecipientName
  • RecipientAddress
  • RecipientCityOrTownStateOrProvinceCountryAndZipOrForeignPostalCode
  • AccountNumber
  • UnemploymentCompensation
  • StateOrLocalIncomeTaxRefundsCreditsOrOffsets
  • Box2AmountIsForTaxYear
  • FederalIncomeTaxWithheld
  • RTAAPayments
  • TaxableGrants
  • AgriculturePayments
  • MarketGain
  • State_Line1
  • State_Line2
  • StateIdentificationNo_Line1
  • StateIdentificationNo_Line2
  • StateIncomeTaxWithheld_Line1
  • StateIncomeTaxWithheld_Line2
  • IsBox2TradeOrBusinessIncome
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

1099-INT Parser

Description

Extract from Form 1099-INT, including payer, recipient, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1099INT_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
  • 2018 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1099int-v1.0-2021-05-27 Stable

None

None

pretrained-1099int-v1.1-2021-12-10 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • PayerName
  • PayerAddressAndPhone
  • PayerTIN
  • RecipientTIN
  • RecipientName
  • RecipientAddress
  • RecipientCityOrTownStateOrProvinceCountryAndZipOrForeignPostalCode
  • AccountNumber
  • PayerRTN
  • InterestIncome
  • EarlyWithdrawalPenalty
  • InterestOnUSSavingsBondsAndTreasObligations
  • FederalIncomeTaxWithheld
  • InvestmentExpenses
  • ForeignTaxPaid
  • ForeignCountryOrUSPossession
  • TaxExemptInterest
  • SpecifiedPrivateActivityBondInterest
  • MarketDiscount
  • BondPremium
  • BondPremiumOnTreasuryObligations
  • BondPremiumOnTaxExemptBond
  • TaxExemptAndTaxCreditBondCUSIPNo
  • State_Line1
  • State_Line2
  • StateIdentificationNo_Line1
  • StateIdentificationNo_Line2
  • StateTaxWithheld_Line1
  • StateTaxWithheld_Line2
  • FatcaFilingRequirement
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

1099-MISC Parser

Description

Extract from Form 1099-MISC, including payer, recipient, amounts, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_1099MISC_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
  • 2018 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1099misc-v1.0-2021-05-27 Stable

None

None

pretrained-1099misc-v1.1-2021-12-10 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • PayerName
  • PayerAddressAndPhone
  • PayerTIN
  • RecipientTIN
  • RecipientName
  • RecipientAddress
  • Rents
  • Royalties
  • OtherIncome
  • FishingBoatProceeds
  • NonemployeeCompensation
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

1099-NEC Parser

Description

Extract from Form 1099-NEC, including payer, recipient, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_1099NEC_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2021 (standard and customized versions)
  • 2020 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1099nec-v1.0-2021-08-11 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • FederalIncomeTaxWithheld
  • NonemployeeCompensation
  • PayersAddress
  • PayersName
  • RecipientAddress
  • RecipientCityStateCountry
  • RecipientName
  • RecipientStreetAddress
  • RecipientTIN
  • StateIncome_Line1
  • StateIncome_Line2
  • StateTaxWithheld_Line1
  • StateTaxWithheld_Line2
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

1099-R Parser

Description

Extract from Form 1099-R, including payer, recipient, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_1099R_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2021 (standard and customized versions)
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1099r-v1.0-2021-08-11 Stable

None

None

pretrained-1099r-v2.0-2022-07-25 Release Candidate
Show fields
  • FormYear
  • PayerFirstName
  • PayerLastName
  • PayerMiddleInitial
  • PayerOrganizationName
  • PayerStreetAddress_Line1
  • PayerStreetAddress_Line2
  • PayerCity
  • PayerState
  • PayerZipcode
  • RecipientFirstName
  • RecipientLastName
  • RecipientMiddleInitial
  • RecipientOrganizationName
  • RecipientCity
  • RecipientState
  • RecipientZipcode
  • RecipientStreetAddress1
  • RecipientStreetAddress2

None

Quality improvements.

Uptraining supported.

Page limit increased from 10 to 15.

Breaking change: PayersName, PayersAddress, RecipientName, ReceptientCityStateCountry, RecipientAddress, RecipientStreetAddress, and EmployerNameAndAddress are no longer part of the output, and they are replaced with additional fields.(for example, PayerStreetAddress_Line1, PayerStreetAddress_Line2, PayerCity, PayerState and PayerZipcode instead of PayersAddress).

LocalTaxWithheld_Line2 is not supported in this version. Please use uptraining function to get the prediction if you are interested.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AmountAllocableToIRRWithin5Years
  • DistributionCode
  • FederalIncomeTaxWithheld
  • GrossDistribution
  • LocalTaxWithheld_Line1
  • LocalTaxWithheld_Line2
  • Other
  • PayersAddress
  • PayersName
  • RecipientAddress
  • RecipientCityStateCountry
  • RecipientName
  • RecipientStreetAddress
  • RecipientTIN
  • StateTaxWithheld_Line1
  • StateTaxWithheld_Line2
  • TaxableAmount
  • TotalEmployeeContributions
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

1065 Parser

Description

Extract from Form 1065, partnership name, address, assets, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_1065_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1065-v1.0-2021-08-11 Stable

None

None

pretrained-1065-v2.0-2022-02-03 Stable

None

None

Quality improvements.

Breaking change: the names of all extracted fields have been renamed from CamelCase to snake_case (for example, end_of_tax_year_cash instead of EndOfTaxYear_Cash).

This change was made to standardize the format of field names across Document AI.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AmountOwed
  • BeginningOfTaxYear_Cash
  • BeginningOfTaxYear_PartnersCapitalAccounts
  • BeginningOfTaxYear_TotalAssets
  • BeginningOfTaxYear_TotalLiabilitiesAndCapital
  • BeginningOfYearBalance
  • BusinessStartDate
  • CityStateCountry
  • EIN
  • EndOfTaxYear_Cash
  • EndOfTaxYear_PartnersCapitalAccounts
  • EndOfTaxYear_TotalAssets
  • EndOfTaxYear_TotalLiabilitiesAndCapital
  • EndOfYearBalance
  • IncomeOrLoss
  • IndividualActive_GeneralPartners
  • IndividualActive_LimitedPartners
  • IndividualPassive_GeneralPartners
  • IndividualPassive_LimitedPartners
  • NetIncomeOrLoss
  • NumberOfScheduleK1
  • OrdinaryBusinessIncomeOrLoss
  • OrdinaryBusinessIncomeOrLoss_ScheduleK
  • Overpayment
  • PartnershipName
  • Partnership_GeneralPartners
  • Partnership_LimitedPartners
  • ScheduleM1_NetIncomeOrLossPerBooks
  • ScheduleM2_NetIncomeOrLossPerBooks
  • SelfEmploymentNetEarningsOrLoss
  • StreetAddress
  • TotalAssets
  • TotalBalanceDue
  • TotalDeductions
  • TotalIncomeOrLoss
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

1120 Parser

Description

Extract from Form 1120, partnership name, address, assets, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_1120_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2021 (pretrained-1120-v3.0-2022-04-26 version only)
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1120-v1.0-2021-08-11 Stable

None

None

pretrained-1120-v2.0-2022-02-03 Stable
Show fields
  • amount_owed
  • bal_begin_tax_yr
  • bal_end_tax_yr
  • begin_of_tax_yr_cash
  • begin_of_tax_yr_total_assets
  • begin_of_tax_yr_total_liab_sh_eq
  • city_state_country
  • credited_to_2021_estimated_tax
  • date_incorporated
  • employer_identification_number
  • end_of_tax_yr_cash
  • end_of_tax_yr_total_assets
  • end_of_tax_yr_total_liab_sh_eq
  • income
  • name
  • net_income_or_loss_per_books
  • over_payment
  • refunded
  • street_address
  • total_assets
  • total_deductions
  • total_income

None

Quality improvements.

Breaking change: the names of all extracted fields have been renamed from CamelCase to snake_case (for example, end_of_tax_year_cash instead of EndOfTaxYear_Cash).

This change was made to standardize the format of field names across Document AI.

pretrained-1120-v3.0-2022-04-26 Release Candidate
Show fields
  • begin_of_tax_yr_mortgages_notes_bonds_payable_below_1_year
  • beginning_date
  • capital_gain_net_income
  • city_state_country_zipcode
  • cost_of_goods_sold
  • depletion
  • depreciation
  • end_of_tax_yr_mortgages_notes_bonds_payable_below_1_year
  • ending_date
  • final_return_checkbox
  • foreign_ownership_no_checkbox
  • foreign_ownership_yes_checkbox
  • gross_income
  • net_gain_or_loss
  • net_operating_loss_deduction
  • other_income
  • other_ownership_no_checkbox_options
  • other_ownership_yes_checkbox_options
  • tax_year
  • taxable_income
  • total_tax_page1
  • travel_and_entertainment

None

Quality improvements.

Added support for Year 2021.

Entity list is in snake_case format similar to pretrained-1120-v2.0-2022-02-03.

Entity city_state_country is renamed into city_state_country_zipcode.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AmountOwed
  • BalanceAtBeginningOfYear
  • BalanceAtEndOfYear
  • BeginningOfTheTaxYear_Cash
  • BeginningOfTheTaxYear_TotalAssets
  • BeginningOfTheTaxYear_TotalLiabilitiesAndShareholdersEquity
  • CityStateCountry
  • CreditedTo2021EstimatedTax
  • DateIncorporated
  • EIN
  • EndOfTheTaxYear_Cash
  • EndOfTheTaxYear_TotalAssets
  • EndOfTheTaxYear_TotalLiabilitiesAndShareholdersEquity
  • Income
  • Name
  • NetIncomeOrLossPerBooks
  • Overpayment
  • Refunded
  • StreetAddress
  • TotalAssets
  • TotalDeductions
  • TotalIncomeOrLoss
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

1120S Parser

Description

Extract from Form 1120S, name, address, assets, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_1120S_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2021 (pretrained-1120s-v2.1-2022-07-22 version only)
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-1120s-v1.0-2021-08-11 Stable

None

None

pretrained-1120s-v2.0-2022-02-03 Stable
Show fields
  • accum_adj_acc_bal_begin_tax_yr
  • accum_adj_acc_bal_end_tax_yr
  • accum_earn_prft_bal_begin_tax_yr
  • accum_earn_prft_bal_end_tax_yr
  • amount_owed
  • begin_of_tax_yr_cash
  • begin_of_tax_yr_total_assets
  • begin_of_tax_yr_total_liab_sh_eq
  • city_state_country
  • credited_to_2021_estimated_tax
  • date_incorporated
  • employer_identification_number
  • end_of_tax_yr_cash
  • end_of_tax_yr_total_assets
  • end_of_tax_yr_total_liab_sh_eq
  • form_year
  • income_or_loss
  • income_or_loss_reconciliation
  • name
  • net_income_or_loss_per_books
  • number_of_shareholders
  • ordinary_biz_income_or_loss
  • ordinary_biz_income_loss_sch_k
  • other_adj_acc_bal_begin_tax_yr
  • other_adj_acc_bal_end_tax_yr
  • over_payment
  • refunded
  • street_address
  • taxable_income_bal_begin_tax_yr
  • taxable_income_bal_end_tax_yr
  • total_assets
  • total_deductions
  • total_income_or_loss

None

Quality improvements.

Breaking change: the names of all extracted fields have been renamed from CamelCase to snake_case (for example, end_of_tax_year_cash instead of EndOfTaxYear_Cash).

This change was made to standardize the format of field names across Document AI.

pretrained-1120s-v2.1-2022-07-22 Release Candidate
Show fields
  • begin_of_tax_yr_accounts_payable
  • begin_of_tax_yr_mortgages_notes_bonds_less_than_a_yr
  • begin_of_tax_yr_other_assets
  • begin_of_tax_yr_other_current_assets
  • begin_of_tax_yr_other_current_liabilities
  • begin_of_tax_yr_tax_exempt_securities
  • begin_of_tax_yr_trade_notes_and_accounts
  • begin_of_tax_yr_us_govt_obligations
  • cost_of_goods_sold
  • depletion
  • depreciation
  • end_of_tax_yr_accounts_payable
  • end_of_tax_yr_mortgages_notes_bonds_less_than_a_yr
  • end_of_tax_yr_other_assets
  • end_of_tax_yr_other_current_assets
  • end_of_tax_yr_other_current_liabilities
  • end_of_tax_yr_tax_exempt_securities
  • end_of_tax_yr_trade_notes_and_accounts
  • end_of_tax_yr_us_govt_obligations
  • tax_year_begin_date
  • tax_year_end_date
  • travel_and_entertainment
  • other_income_or_loss

None

Quality improvements and supporting new fields.

Added support for year 2021.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AccumulatedAdjustmentsAccount_BalanceAtBeginningOfTaxYear
  • AccumulatedAdjustmentsAccount_BalanceAtEndOfTaxYear
  • AccumulatedEarningsAndProfits_BalanceAtBeginningOfTaxYear
  • AccumulatedEarningsAndProfits_BalanceAtEndOfTaxYear
  • AmountOwed
  • BeginningOfTaxYear_Cash
  • BeginningOfTaxYear_TotalAssets
  • BeginningOfTaxYear_TotalLiabilitiesAndShareHoldersEquity
  • CityStateCountry
  • CreditedTo2021EstimatedTax
  • DateIncorporated
  • EIN
  • EndOfTaxYear_Cash
  • EndOfTaxYear_TotalAssets
  • EndOfTaxYear_TotalLiabilitiesAndShareHoldersEquity
  • IncomeOrLoss
  • IncomeOrLossReconciliation
  • Name
  • NetIncomeOrLossPerBooks
  • NumberOfShareholders
  • OrdinaryBusinessIncomeOrLoss
  • OrdinaryBusinessIncomeOrLoss_ScheduleBK
  • OtherAdjustmentsAccount_BalanceAtBeginningOfTaxYear
  • OtherAdjustmentsAccount_BalanceAtEndOfTaxYear
  • Overpayment
  • Refunded
  • ShareholdersUndistributedTaxableIncomePreviouslyTaxed_BalanceAtBeginningOfTaxYear
  • ShareholdersUndistributedTaxableIncomePreviouslyTaxed_BalanceAtEndOfTaxYear
  • StreetAddress
  • TotalAssets
  • TotalDeductions
  • TotalIncomeOrLoss
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

Bank Statement Parser

Description

Extract from bank statements including name, account, transactions, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API BANK_STATEMENT_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-bankstatement-v1.0-2021-08-08 Stable

None

None

pretrained-bankstatement-v1.1-2021-08-13 Stable

None

None

pretrained-bankstatement-v2.0-2021-12-10 Stable

None

None

pretrained-bankstatement-v3.0-2022-05-16 Stable

None

None

This version assumes that the input file contains a single bank statement. Unlike the default version, this version does not check the input file for bank statements and will not return an error if no bank statements are found. If your input document contains multiple bank statements, use the Lending Document Splitter & Classifier for splitting before sending it to this processor.
pretrained-bankstatement-v4.0-2023-07-31 Release Candidate

None

None

pretrained-bankstatement-v5.0-2023-12-06 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 30
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • account_number
  • account_type
  • bank_address
  • bank_name
  • client_address
  • client_name
  • ending_balance
  • starting_balance
  • statement_date
  • statement_end_date
  • statement_start_date
  • table_item
    • table_item/transaction_deposit
    • table_item/transaction_deposit_date
    • table_item/transaction_deposit_description
    • table_item/transaction_withdrawal
    • table_item/transaction_withdrawal_date
    • table_item/transaction_withdrawal_description
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • bank_address
  • bank_name
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • ending_balance
  • starting_balance
  • statement_date
  • statement_end_date
  • statement_start_date
  • table_item/transaction_deposit
  • table_item/transaction_deposit_date
  • table_item/transaction_withdrawal
  • table_item/transaction_withdrawal_date
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

HOA Statement Parser

Description

Extract from Homeowner Association(HOA) statements including name, address, due amount, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API HOA_STATEMENT_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file.
  • Extracted 'other_name' and 'other_address' represent HOA related names and addresses (such as property management company, HOA organization, payment service company, etc.) other than explictly extracted property_owner_name/address or property_address.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-hoa-statement-v1.0-2021-12-08 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 50
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • due_amount
  • due_date
  • frequency
  • late_fee
  • other_address
  • other_name
  • paid_amount
  • paid_date
  • paid_through_amount
  • paid_through_date
  • property_address
  • property_owner_address
  • property_owner_name
  • statement_date
  • table_item
    • table_item/transaction_charge
    • table_item/transaction_date
    • table_item/transaction_description
    • table_item/transaction_payment
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • other_address
  • other_name
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

HUD-92900B Parser

Description

Extract from Form HUD-92900B dates and signature existence.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_HUD92900B_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2019 (standard version only)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-hud92900b-v1.0-2021-09-16 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • borrower1_signature_date_exists
  • borrower1_signature_exists
  • borrower2_signature_date_exists
  • borrower2_signature_exists
  • borrower3_signature_date_exists
  • borrower3_signature_exists
  • borrower4_signature_date_exists
  • borrower4_signature_exists
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

Lending Document Splitter & Classifier

Description

Identify documents in a large file and classify known lending document types.

Mortgage application packages and other lending documents often contain multiple documents (such as 1040 tax forms, W2, bank statements, etc.) in a single file. Lending document splitter allows you to programmatically split these combined lending documents on logical boundaries. The split files are then classified based on the document type, so that the appropriate extraction model can be applied to each file.

Category Specialized
Solution type Lending
Functions OCR, Classification, Splitting
Release stage General availability
Access status Limited [3]
Type in API LENDING_DOCUMENT_SPLIT_PROCESSOR
Notes
  • The splitter is not designed to split logical documents that are over 30 pages long. Logical documents that are more than 30 pages long (e.g. a 40-page bank statement) may be split into two or more docs and classified separately.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-lending-document-split-v1.0-2021-12-08 Stable

None

None

pretrained-lending-document-split-v2.0-2021-12-09 Stable
Show fields
  • 1005_1996
  • 1040_2021[1]
  • 1040nr[1]
  • 1040nr_2018
  • 1040nr_2019
  • 1040nr_2020
  • 1040nr_2021[1]
  • 1040sb[1]
  • 1040sb_2018
  • 1040sb_2019
  • 1040sb_2020
  • 1040sb_2021[1]
  • 1040sc_2021[1]
  • 1040sd[1]
  • 1040sd_2018
  • 1040sd_2019
  • 1040sd_2020
  • 1040sd_2021[1]
  • 1040se_2021[1]
  • 1040sr[1]
  • 1040sr_2018
  • 1040sr_2019
  • 1040sr_2020
  • 1040sr_2021[1]
  • 1065_2021
  • 1076_2016
  • 1099div_2021[1]
  • 1099g_2021[1]
  • 1099int_2021[1]
  • 1099misc_2021[1]
  • 1099nec_2018[1]
  • 1099nec_2019[1]
  • 1099nec_2021
  • 1099r_2021[1]
  • 1099ssa_2021
  • 1120_2021[1]
  • 1120s_2021[1]
  • 1_4_Family_Rider_3170
  • 3108_Adjustable_Rate_Rider
  • 3140_Condominium_Rider
  • 3190_Balloon_Rider
  • 3890_Second_Home_Rider
  • 4506_T[1]
  • 4506_T_2018[1]
  • 4506_T_2019[1]
  • 4506_T_2020[1]
  • 4506_T_2021
  • 4506_T_EZ[1]
  • 4506_T_EZ_2018[1]
  • 4506_T_EZ_2019
  • 4506_T_EZ_2020[1]
  • 4506_T_EZ_2021
  • account_statement_investment_and_retirement
  • appraisal_ucdp_ssr
  • dhs_flood_certification
  • f11_12956_2017[1]
  • hud_54114
  • hud_92051
  • hud_92541
  • hud_92544
  • hud_92800
  • hud_92900a
  • hud_92900b
  • hud_92900lt
  • hud_92900ws
  • mortgage_statements
  • property_insurance
  • pud_rider
  • revocable_trust_rider
  • ssa_89[1]
  • ssa_89_2018[1]
  • ssa_89_2019[1]
  • ssa_89_2020
  • ssa_89_2021
  • ucc_financing_statement
  • usda_ad_3030
  • vba_26_0551_2004
  • vba_26_8923_2021
  • w2_2021[1]
  • w9_2019[1]
  • w9_2020[1]
  • w9_2021[1]

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 1250
Uptraining
Human-in-the-Loop[2]
Document types identified
Show types

This splitter can identify and classify the following types of documents and form:

  • 1003 - Legacy Form (standard and customized versions)
    • Return type(s): 1003[1], 1003_2009
  • 1040 - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1040[1], 1040_2018, 1040_2019, 1040_2020[1]
  • 1040 Schedule C - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1040sc[1], 1040sc_2018[1], 1040sc_2019, 1040sc_2020
  • 1040 Schedule E - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1040se[1], 1040se_2018[1], 1040se_2019, 1040se_2020
  • 1065 - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1065[1], 1065_2018[1], 1065_2019, 1065_2020
  • 1099-DIV - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1099div[1], 1099div_2018, 1099div_2019, 1099div_2020
  • 1099-G - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1099g[1], 1099g_2018[1], 1099g_2019, 1099g_2020
  • 1099-INT - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1099int[1], 1099int_2018, 1099int_2019, 1099int_2020
  • 1099-MISC - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1099misc[1], 1099misc_2018, 1099misc_2019, 1099misc_2020
  • 1099-NEC - 2020 (standard and customized versions)
    • Return type(s): 1099nec[1], 1099nec_2020
  • 1099-R - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1099r[1], 1099r_2018, 1099r_2019, 1099r_2020
  • 1120 - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1120[1], 1120_2018[1], 1120_2019, 1120_2020
  • 1120S - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1120s[1], 1120s_2018[1], 1120s_2019, 1120s_2020
  • Bank Statement
    • Return type(s): account_statement_bank
  • Pay Slip
    • Return type(s): payslip
  • SSA-1099 - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): 1099ssa[1], 1099ssa_2018[1], 1099ssa_2019, 1099ssa_2020
  • US Driver License
    • Return type(s): US_Driver_License
  • US Pasport
    • Return type(s): US_Passport
  • W2 - 2018, 2019, 2020 (standard and customized versions)
    • Return type(s): w2[1], w2_2018, w2_2019, w2_2020
  • W9 - Rev. 10-2018, Rev. 11-2017
    • Return type(s): w9[1], w9_2017, w9_2018
  • If the splitter cannot identify the type of the document, it returns other.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us
More information Document splitters behavior

Mortgage Statement Parser

Description

Extract from mortgage statements including name, address, due amount, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API MORTGAGE_STATEMENT_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-mortgage-statement-v1.0-2021-10-17 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 50
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • borrower_address
  • borrower_name
  • due_date
  • fees_due
  • frequency
  • insurance_escrow_due
  • interest_due
  • interest_rate
  • loan_number
  • loan_type
  • maturity_date
  • others_due
  • past_due_amount
  • principal_balance
  • principal_due
  • property_address
  • regular_payment_amount
  • servicer_address
  • servicer_name
  • statement_date
  • table_item
    • table_item/description
    • table_item/effective_date
    • table_item/escrow_amount
    • table_item/fees_amount
    • table_item/interest_amount
    • table_item/others_amount
    • table_item/principal_amount
    • table_item/total_amount
    • table_item/total_charged
    • table_item/total_paid
    • table_item/transaction_date
    • table_item/unapplied_amount
  • tax_escrow_due
  • total_due
  • total_escrow_due
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • servicer_address
  • servicer_name
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

Pay Slip Parser

Description

Extract from pay slips, including name, business, amounts, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API PAYSTUB_PROCESSOR
Notes
  • If the multi-page input document contains more than one valid pay slips, the processor extracts entities from only the first valid pay slip. If no pay slips are found in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-paystub-v1.0-2021-03-19 Stable

None

None

pretrained-paystub-v1.1-2021-08-13 Stable
Show fields
  • net_pay
  • net_pay_ytd
  • employee_account_number

None

Quality improvement and new fields support;
pretrained-paystub-v1.2-2021-12-10 Stable

None

None

pretrained-paystub-v2.0-2022-05-17 Release Candidate
Show fields
  • deduction_item
  • deduction_item/deduction_type
  • deduction_item/deduction_this_period
  • deduction_item/deduction_ytd
  • direct_deposit_item
  • direct_deposit_item/direct_deposit
  • direct_deposit_item/employee_account_number
  • earning_item
  • earning_item/earning_type
  • earning_item/earning_rate
  • earning_item/earning_hours
  • earning_item/earning_this_period
  • earning_item/earning_ytd
  • page_number
  • tax_item
  • tax_item/tax_type
  • tax_item/tax_this_period
  • tax_item/tax_ytd
  • federal_additional_tax
  • federal_allowance
  • federal_marital_status
  • state_additional_tax
  • state_allowance
  • state_marital_status

None

This version assumes that the input file contains a single pay slip. Unlike the default version, this version does not check the input file for pay slips and will not return an error if no pay slips are found. If your input document contains multiple pay slips, use the Lending Document Splitter & Classifier for splitting before sending it to this processor.

Quality improvement, new fields support and new schema. Bonus, Commissions, Holiday, Overtime, Regular Pay and Vacation are now part of earning_item/earning_this_period, and their year-to-date versions are in earning_item/earning_ytd. Direct Deposit and Employee Account Number are now nested under direct_deposit_item.

Async page limit is 10.

pretrained-paystub-v2.0-2022-07-22 Stable

None

None

Quality improvement and uptraining enhancements.

pretrained-paystub-v3.0-2023-12-06 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 50
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • bonus
  • bonus_ytd
  • commissions
  • commissions_ytd
  • direct_deposit
  • employee_account_number (Added in "pretrained-paystub-v1.1-2021-08-13")
  • employee_address
  • employee_name
  • employer_address
  • employer_name
  • end_date
  • gross_earnings
  • gross_earnings_ytd
  • holiday
  • holiday_ytd
  • net_pay (Added in "pretrained-paystub-v1.1-2021-08-13")
  • net_pay_ytd (Added in "pretrained-paystub-v1.1-2021-08-13")
  • overtime
  • overtime_ytd
  • pay_date
  • regular_pay
  • regular_pay_ytd
  • ssn
  • start_date
  • vacation
  • vacation_ytd
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • employer_address
  • employer_name
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • bonus
  • bonus_ytd
  • commissions
  • commissions_ytd
  • direct_deposit
  • end_date
  • gross_earnings
  • gross_earnings_ytd
  • holiday
  • holiday_ytd
  • net_pay
  • net_pay_ytd
  • overtime
  • overtime_ytd
  • pay_date
  • regular_pay
  • regular_pay_ytd
  • start_date
  • vacation
  • vacation_ytd
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

Retirement/Investment Statement Parser

Description

Extract from Retirement/Investment statements including name, address, due amount, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API RETIREMENT_INVESTMENT_STATEMENT_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-retirement-investment-statement-v1.0-2021-12-03 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 30
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • account_number
  • account_type
  • client_address
  • client_name
  • deposit_amount
  • deposit_source
  • ending_balance
  • fbo_itf_or_trust_indicator
  • financial_institution_address
  • financial_institution_name
  • starting_balance
  • statement_end_date
  • statement_start_date
  • withdrawal_amount
  • withdrawal_source
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • financial_institution_address
  • financial_institution_name
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • deposit_amount
  • ending_balance
  • starting_balance
  • statement_end_date
  • statement_start_date
  • withdrawal_amount
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

SSA-89 Parser

Description

Extract from Form SSA-89, including name, address, SSN, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_SSA89_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-ssa89-v1.0-2021-09-16 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • agents_address
  • agents_name
  • company_address
  • company_name
  • date_of_birth
  • date_signed_exists
  • initials_exists
  • no_of_days_after_signature_that_consent_is_valid
  • reason_for_authorizing_consent_other_value
  • printed_name
  • reason_for_authorizing_consent
  • relationship
  • signature_exists
  • social_security_number
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

SSA-1099 Parser

Description

Extract from Form SSA-1099 including name, address, SSN, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_SSA1099_PROCESSOR
Notes
  • This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-ssa1099-v1.0-2021-08-09 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • Address
  • BenefitsPaid
  • BenefitsRepaidToSSA
  • DescriptionOfAmountInBox3
  • DescriptionOfAmountInBox4
  • Name
  • NetBenefits
  • SSN
  • VoluntaryFederalIncomeTaxWithheld
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

VBA26-0551 Parser

Description

Extract from Form VBA26-0551, coborrower signature, veteran signature, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage Preview
Access status Limited [3]
Type in API FORM_VBA26_0551_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2004 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-vba26-0551-v1.0-2021-09-16 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • delinquent_default_any_debt_federal_govt_checkbox
  • past5year_loan_foreclosure_liue_judgment_checkbox
  • signature_of_coborrower_date_exist
  • signature_of_coborrower_exist
  • signature_of_veteran_date_exist
  • signature_of_veteran_exist
Uptraining
Human-in-the-Loop[2]
Supported Regions
  • eu
  • us

W2 Parser

Description

Extract from Form W2, including employee, employer, wages, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_W2_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • 2020 (standard and customized versions)
  • 2019 (standard and customized versions)
  • 2018 (standard and customized versions)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-w2-v1.0-2020-10-01 Stable

None

None

pretrained-w2-v1.1-2022-01-27 Stable

None

None

pretrained-w2-v1.2-2022-01-28 Stable
Show fields
  • AllocatedTips
  • ControlNumber
  • DependentCareBenefits
  • EIN
  • EmployeeAddress
  • EmployeeName
  • EmployerNameAndAddress
  • EmployerStateIdNumber_Line1
  • FederalIncomeTaxWithheld
  • FormYear
  • LocalIncomeTax_Line1
  • LocalityName_Line1
  • LocalWagesTipsEtc_Line1
  • MedicareTaxWithheld
  • MedicareWagesAndTips
  • NonqualifiedPlans
  • SocialSecurityTaxWithheld
  • SocialSecurityTips
  • SocialSecurityWages
  • SSN
  • State_Line1
  • StateIncomeTax_Line1
  • StateWagesTipsEtc_Line1
  • WagesTipsOtherCompensation

None

Quality improvements and supporting new fields; does not include splitter.

pretrained-w2-v2.0-2022-03-30 Release Candidate
Show fields
  • AllocatedTips
  • ControlNumber
  • DependentCareBenefits
  • EIN
  • EmployeeAddress_AdditionalStreetAddressOrPostalBox
  • EmployeeAddress_City
  • EmployeeAddress_State
  • EmployeeAddress_StreetAddressOrPostalBox
  • EmployeeAddress_Zip
  • EmployeeName_FirstName
  • EmployeeName_LastName
  • EmployeeName_MiddleNameOrInitial
  • EmployerAddress_AdditionalStreetAddressOrPostalBox
  • EmployerAddress_City
  • EmployerAddress_State
  • EmployerAddress_StreetAddressOrPostalBox
  • EmployerAddress_Zip
  • EmployerName
  • EmployerStateIdNumber_Line1
  • FederalIncomeTaxWithheld
  • FormYear
  • LocalIncomeTax_Line1
  • LocalWagesTipsEtc_Line1
  • LocalityName_Line1
  • MedicareTaxWithheld
  • MedicareWagesAndTips
  • NonqualifiedPlans
  • SSN
  • SocialSecurityTaxWithheld
  • SocialSecurityTips
  • SocialSecurityWages
  • StateIncomeTax_Line1
  • StateWagesTipsEtc_Line1
  • State_Line1
  • WagesTipsOtherCompensation
  • a_Code
  • a_Value
  • b_Code
  • b_Value
  • c_Code
  • c_Value
  • d_Code
  • d_Value

None

Quality improvements and support for box 12 fields and fine-grained predictions of EmployeeName, EmployeeAddress, and EmployerNameAndAddress, all of which are no longer part of the output and are replaced with additional fields.

pretrained-w2-v2.1-2022-06-08 Stable
Show fields
  • AllocatedTips
  • ControlNumber
  • DependentCareBenefits
  • EIN
  • EmployeeAddress_AdditionalStreetAddressOrPostalBox
  • EmployeeAddress_City
  • EmployeeAddress_State
  • EmployeeAddress_StreetAddressOrPostalBox
  • EmployeeAddress_Zip
  • EmployeeName_FirstName
  • EmployeeName_LastName
  • EmployeeName_MiddleNameOrInitial
  • EmployeeName_Suffix
  • EmployerAddress_AdditionalStreetAddressOrPostalBox
  • EmployerAddress_City
  • EmployerAddress_State
  • EmployerAddress_StreetAddressOrPostalBox
  • EmployerAddress_Zip
  • EmployerName
  • EmployerStateIdNumber_Line1
  • FederalIncomeTaxWithheld
  • FormYear
  • LocalIncomeTax_Line1
  • LocalWagesTipsEtc_Line1
  • LocalityName_Line1
  • MedicareTaxWithheld
  • MedicareWagesAndTips
  • NonqualifiedPlans
  • SSN
  • SocialSecurityTaxWithheld
  • SocialSecurityTips
  • SocialSecurityWages
  • StateIncomeTax_Line1
  • StateWagesTipsEtc_Line1
  • State_Line1
  • WagesTipsOtherCompensation
  • a_Code
  • a_Value
  • b_Code
  • b_Value
  • c_Code
  • c_Value
  • d_Code
  • d_Value

None

Similar to version pretrained-w2-v2.0-2022-03-30 with further quality enhancements and introducing one more entity EmployeeName_Suffix.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • ControlNumber
  • EIN
  • EmployeeAddress
  • EmployeeName
  • EmployerNameAndAddress
  • FederalIncomeTaxWithheld
  • MedicareTaxWithheld
  • MedicareWagesAndTips
  • SSN
  • SocialSecurityTaxWithheld
  • SocialSecurityWages
  • WagesTipsOtherCompensation
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • EmployerNameAndAddress
  • EIN
Uptraining
Human-in-the-Loop[2]
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

W9 Parser

Description

Extract from Form W9 including name, address, TIN, etc.

Category Specialized
Solution type Lending
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API FORM_W9_PROCESSOR
Notes
  • If a page of a multi-page input file is the correct document type and one of the supported versions, the processor performs entity extraction on the first supported document. If the processor doesn't find any applicable documents in the input file, the processor returns an error message.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Supported form/versions
  • Form (Rev. 10-2018, Rev. 11-2017)
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-w9-v1.0-2020-09-25 Stable

None

None

pretrained-w9-v1.1-2021-12-10 Stable

None

None

pretrained-w9-v1.2-2022-01-27 Stable

None

None

Handles documents with variations other than the standard template and does not include splitter and classifier model (that has its own service and can be called seperately);
pretrained-w9-v2.0-2022-06-23 Release Candidate

None

None

Quality improvements.

Sync page limit is 10.

Breaking change: the names of all extracted fields have been renamed from CamelCase to snake_case (for example, business_name instead of BusinessName).

This change was made to standardize the format of field names across Document AI.

This processor assumes the input file contains the supported document from the beginning and will not classify or split the input file. If your input file does not meet this assumption, please run the Lending Document Splitter & Classifier first and preprocess the input file.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 15
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • AccountNumbers
  • Address
  • BusinessName
  • CityStateZip
  • EIN
  • ExemptFatcaCode
  • ExemptPayeeCode
  • FederalTaxClassification
  • FederalTaxClassificationOther
  • FormRevisionDate
  • HasSignature
  • HasSignatureDate
  • LlcTaxClassification
  • Name
  • SSN

Breaking change for Google Release Candidate 2022-06-23: the names of all extracted fields have been renamed from CamelCase to snake_case (for example, business_name instead of BusinessName). This change was made to standardize the format of field names across Document AI.

Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • Address
  • BusinessName
  • EIN
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

Expense Parser

Description

Extract text and values from expense documents such as expense date, supplier name, total amount, and currency.

Category Specialized
Solution type Procurement
Functions OCR, Entity Extraction
Release stage General availability
Access status Public
Type in API EXPENSE_PROCESSOR
Supported languages
Full list of languages
Language Name BCP 47 Tag Script Handwriting supported
German de Latn
English en Latn
Spanish es Latn
French fr Latn
Japanese ja Jpan
Dutch nl Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-expense-v1.1-2021-04-09 Stable

None

None

Launched in April 2021. Deprecation is planned soon.
pretrained-expense-v1.2-2022-02-18 Stable

None

None

pretrained-expense-v1.3-2022-07-15 Stable
Show fields
  • credit_card_last_four_digits
  • line_item/quantity
  • payment_type
  • ja: Japanese
Support for hotel and car rental folios.
pretrained-expense-v1.4-2022-11-18 Release Candidate
Show fields
  • traveler_name
  • reservation_id
  • line_item/transaction_date
  • ja: Japanese
  • it: Italian
  • pt: Portuguese (Portugal & Brazil)
Performance improvements and support for uptraining. Maximum pages (online/synchronous requests) limit has been increased to 15.

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 10
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • credit_card_last_four_digits
  • currency
  • end_date
  • net_amount
  • payment_type
  • purchase_time
  • receipt_date
  • start_date
  • supplier_address
  • supplier_city
  • supplier_name
  • tip_amount
  • total_amount
  • total_tax_amount
  • line_item
    • line_item/amount
    • line_item/description
    • line_item/product_code
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • supplier_address
  • supplier_name
  • supplier_phone
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • currency
  • total_amount
  • total_tax_amount
  • net_amount
  • receipt_date
  • purchase_time
  • start_date
  • end_date
  • line_item/amount
  • line_item/payment_date
  • line_item/payment_amount
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-southeast1
  • australia-southeast1
  • eu
  • northamerica-northeast1
  • us

Invoice Parser

Description

Extract text and values from invoices such as invoice number, supplier name, invoice amount, tax amount, invoice date, due date.

The invoice Parser extracts both header and line item fields, such as invoice number, supplier name, invoice amount, tax amount, invoice date, due date, and line item amounts.

Category Specialized
Solution type Procurement
Functions OCR, Entity Extraction
Release stage General availability
Access status Public
Type in API INVOICE_PROCESSOR
Supported languages
Full list of languages
Language Name BCP 47 Tag Script Handwriting supported
German de Latn
English en Latn
Spanish es Latn
Estonian et Latn
French fr Latn
Italian it Latn
Latvian lv Latn
Lithuanian lt Latn
Dutch nl Latn
Portuguese (Portugal & Brazil) pt Latn
Romanian ro Latn
Swedish sv Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-invoice-v1.1-2021-04-09 Stable

None

None

pretrained-invoice-v1.2-2022-02-18 Stable

None

None

Deprecation is planned soon.
pretrained-invoice-v1.3-2022-07-15 Stable

None

  • it: Italian
  • pt: Portuguese (Portugal & Brazil)
  • ro: Romanian
  • sv: Swedish
  • et: Estonian
  • lv: Latvian
  • lt: Lithuanian
Uptrainable processor version. Maximum pages (online/synchronous requests) has been increased to 15.
pretrained-invoice-v1.4-2022-10-21 Release Candidate

None

None

Uptrainable processor version. Maximum pages (online/synchronous requests) has been increased to 15.
pretrained-invoice-v1.5-2023-09-15 Release Candidate

None

None

pretrained-invoice-v2.0-2023-12-06 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 200
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • amount_paid_since_last_invoice
  • carrier
  • currency
  • currency_exchange_rate
  • delivery_date
  • due_date
  • freight_amount
  • invoice_date
  • invoice_id
  • line_item
    • line_item/amount
    • line_item/description
    • line_item/product_code
    • line_item/purchase_order
    • line_item/quantity
    • line_item/unit
    • line_item/unit_price
  • net_amount
  • payment_terms
  • purchase_order
  • receiver_address
  • receiver_email
  • receiver_name
  • receiver_phone
  • receiver_tax_id
  • receiver_website
  • remit_to_address
  • remit_to_name
  • ship_from_address
  • ship_from_name
  • ship_to_address
  • ship_to_name
  • supplier_address
  • supplier_email
  • supplier_iban
  • supplier_name
  • supplier_payment_ref
  • supplier_phone
  • supplier_registration
  • supplier_tax_id
  • supplier_website
  • total_amount
  • total_tax_amount
  • vat
    • vat/amount
    • vat/category_code
    • vat/tax_amount
    • vat/tax_rate
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • supplier_address
  • supplier_name
  • supplier_phone
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • amount_paid_since_last_invoice
  • currency
  • currency_exchange_rate
  • delivery_date
  • due_date
  • freight_amount
  • invoice_date
  • net_amount
  • total_amount
  • total_tax_amount
  • line_item/amount
  • line_item/quantity
  • line_item/unit_price
  • vat/amount
  • vat/tax_amount
  • vat/tax_rate
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • asia-south1
  • asia-southeast1
  • australia-southeast1
  • eu
  • northamerica-northeast1
  • us

Procurement Document Splitter & Classifier

Description

Allows you to programmatically split these combined procurement documents on logical boundaries.

Procurement document splitter allows you to take different procurement documents grouped in a single file and programmatically split the documents on logical boundaries. The split files are then classified based on the document type, so that the appropriate extraction model can be applied to each file.

Category Specialized
Solution type Procurement
Functions OCR, Classification, Splitting
Release stage General availability
Access status Limited [3]
Type in API PROCUREMENT_DOCUMENT_SPLIT_PROCESSOR
Notes
  • The splitter is not designed to split logical documents that are over 30 pages long. Logical documents that are more than 30 pages long (e.g. a 40-page bank statement) may be split into two or more docs and classified separately.
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-procurement-splitter-v1.1-2021-04-09 Stable

None

None

pretrained-procurement-splitter-v1.2-2022-08-19 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 15
Maximum pages (batch/offline/asynchronous requests): 1250
Uptraining
Human-in-the-Loop[2]
Document types identified
Show types

This splitter can identify and classify the following types of documents and form:

  • Utility statement: A bill or receipt issued by an utility company (telecommunications, gas, electric, cable service) that shows the amount owed by the customer for the services provided. This may also show the previous payments made by the customer for current or prior services.
    • Return type(s): utility_statement
  • Debit note: A document issued by a business stating a monetary amount a client owes to the business.
    • Return type(s): debit_note
  • Credit note: A document issued by a business that needs to provide a client with a discount or a refund, or to correct a previous invoicing error.
    • Return type(s): credit_note
  • Credit Card Slip: A document that shows a payment made by credit card. It typically includes the total charge amount, a tip amount (mostly US documents), and a total payment. Tip amount and total payment are usually handwritten. This doc is relevant for expense processing. It is not a suitable proof of expense in expense processing.
    • Return type(s): credit_card_slip
  • Restaurant statement: A document issued by a restaurant to a customer itemizing the specific items consumed, the taxes, total amount, tips, and amount paid.
    • Return type(s): restaurant_statement
  • Air travel statement: A document issued by an airline to a customer itemizing the specific flight and non-flight charges, and the amount paid (if available).
    • Return type(s): air_travel_statement
  • Hotel statement: A document issued by a hotel to a customer itemizing the specific charges related to a hotel stay, and the amount paid (if available).
    • Return type(s): hotel_statement
  • Car rental statement: A document issued by a car rental company to a customer itemizing the specific charges related to a car rental, and the amount paid (if available).
    • Return type(s): car_rental_statement
  • Ground transportation statement: A document issued by a ground transportation company (ride sharing, train/subway) to a customer itemizing the specific charges related to a trip, and the amount paid (if available).
    • Return type(s): ground_transportation_statement
  • Invoice statement: A document sent by the seller to the customer that requests payments for products or services and (for the purpose of our taxonomy) is not covered by any other document type definition.
    • Return type(s): invoice_statement
  • Receipt statement: A document that shows proof of payment which confirms that a customer has received the goods and services they paid a business for. Conversely, this can be a document showing the business was compensated for the goods or services they sold to a customer and (for the purpose of our taxonomy) is not covered by any other document type definition.
    • Return type(s): receipt_statement
  • If the splitter cannot identify the type of the document, it returns other.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us
More information Document splitters behavior

Purchase Order Parser

Description

Extract text and values from purchase orders such as shipping address, purchase order id, purchase order date, total amount, delivery date.

Category Specialized
Solution type Procurement
Functions OCR, Entity Extraction
Release stage Preview
Access status Public
Type in API PURCHASE_ORDER_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-purchase-order-v1.1-2022-06-17 Stable

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 200
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • currency
  • delivery_date
  • line_item
    • line_item/amount
    • line_item/description
    • line_item/product_code
    • line_item/purchase_order_id
    • line_item/quantity
    • line_item/receiver_reference
    • line_item/unit_of_measure
    • line_item/unit_price
  • payment_terms
  • purchase_order_date
  • purchase_order_id
  • receiver_name
  • ship_to_address
  • ship_to_name
  • total_amount
Enriched fields

You can find more information in the Enrichment & normalization page.

Full list of enriched fields
  • receiver_name
  • ship_to_address
  • ship_to_name
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • currency
  • delivery_date
  • total_amount
  • line_item/amount
  • line_item/quantity
  • line_item/unit_price
  • purchase_order_date
  • total_amount
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Supported Regions
  • eu
  • us

Utility Parser

Description

Extract text and values from utility bills such as supplier name and previous paid amount.

Category Specialized
Solution type Procurement
Functions OCR, Entity Extraction
Release stage General availability
Access status Limited [3]
Type in API UTILITY_PROCESSOR
Supported languages
Language Name BCP 47 Tag Script Handwriting supported
English en Latn
Processor versions
Version ID Release Channel Additional fields detected Additional languages supported Description
pretrained-utility-v1.1-2021-04-09 Stable

None

None

pretrained-utility-v1.2-2022-12-15 Release Candidate

None

None

For more information, see Managing processor versions.

Quotas and limits
Maximum pages (online/synchronous requests): 10
Maximum pages (batch/offline/asynchronous requests): 200
Fields detected in the earliest version

You can also find this information in the Field detected page.

Full list of fields
  • adjusted_amount
  • amount_due
  • balance_transfer_amount
  • carrier
  • currency
  • currency_exchange_rate
  • delivery_date
  • deposit_credited_amount
  • due_date
  • freight_amount
  • invoice_date
  • invoice_id
  • late_fee_amount
  • line_item
    • line_item/amount
    • line_item/description
    • line_item/frequency
    • line_item/product_code
    • line_item/purchase_order
    • line_item/quantity
    • line_item/service_address
    • line_item/service_end_date
    • line_item/service_id_1
    • line_item/service_id_2
    • line_item/service_start_date
    • line_item/supplier_account_number
    • line_item/tax_amount
    • line_item/unit_number
    • line_item/unit_of_measure
    • line_item/unit_price
    • line_item/usage
  • net_amount
  • payment_terms
  • prior_amount_due
  • prior_paid_amount
  • purchase_order
  • receiver_address
  • receiver_email
  • receiver_name
  • receiver_phone
  • receiver_tax_id
  • receiver_website
  • reclaimed_water
  • remit_to_address
  • remit_to_name
  • service
    • service/service_end_date
    • service/service_id
    • service/service_start_date
    • service/unit_of_measure
    • service/usage
  • service_address
  • service_end_date
  • service_id
  • service_start_date
  • ship_from_address
  • ship_from_name
  • ship_to_address
  • ship_to_name
  • supplier_account_number
  • supplier_address
  • supplier_email
  • supplier_iban
  • supplier_name
  • supplier_payment_ref
  • supplier_phone
  • supplier_registration
  • supplier_tax_id
  • supplier_website
  • tampering
  • total_amount
  • total_tax_amount
  • usage
  • vat
    • vat/amount
    • vat/category_code
    • vat/tax_amount
    • vat/tax_rate
Normalized fields

You can find more information in the Enrichment & normalization page.

Full list of normalized fields
  • adjusted_amount
  • amount_due
  • balance_transfer_amount
  • currency
  • currency_exchange_rate
  • delivery_date
  • due_date
  • invoice_date
  • late_fee_amount
  • line_item/amount
  • line_item/quantity
  • line_item/tax_amount
  • line_item/unit_price
  • net_amount
  • prior_amount_due
  • prior_paid_amount
  • total_amount
  • total_tax_amount
Uptraining
Human-in-the-Loop[2]
Labeling Instructions Open in new window.
Sample Input File Open in new window.
Sample Output Open in new window.
Supported Regions
  • eu
  • us

Footnotes

[1] The corresponding parser for this form does not support this doc type. This means that the splitter can identify and classify documents of this type, but Document AI does not provide a parser to extract information.

[2]

To request API access, fill out and submit the Contract AI limited access customer request form. The form requests information about you, your company, and your use case. Note that a Google Cloud Project ID is required for access. To create a new Google Cloud project, or identify your existing project's Project ID see the following instructions.

After you submit the form, the Document AI team will review your request to ensure you meet the criteria for access. If approved, you will receive an email with instructions on how to access and use this feature.

[3]

To request API access, fill out and submit the Document AI limited access customer request form. The form requests information about you, your company, and your use case. Note that a Google Cloud Project ID is required for access. To create a new Google Cloud project, or identify your existing project's Project ID see the following instructions.

After you submit the form, the Document AI team will review your request to ensure you meet the criteria for access. If approved, you will receive an email with instructions on how to access and use this feature.

[4] Identity Document Proofing works to extract and evaluate information from ID documents that contributes to identifying whether the input image represents an authentic ID.

At Google Cloud, we prioritize helping customers safely develop and implement AI solutions, and Identity Proofing has been developed in accordance with Google's AI Principles.

Based on Google's AI Principles and current product design, we strongly recommend using caution and carefully evaluating the potential benefits and risks of using Identity Document Proofing for the following:

  • Decision-making without a human in the loop for predictions that can impact human rights.
  • In sensitive domains including but not limited to employment, access to public services, healthcare, and safety-critical contexts.

[5] Always use Identity Proofing as part of your broader identity detection process and workflow. It is important that you have a human reviewer in your workflow to verify whether the predicted signals are accurate. The Identity Proofing processor isn't meant to replace human review of IDs in a workflow, but rather to assist human reviewers in validating ID documents. The Identity Proofing processor shouldn't be used as an automated decision tool to determine whether an ID is valid. With human review, customers can achieve higher document processing accuracy and help businesses evaluate predictions using purpose-built tools to enable those reviews.

Make sure that you review regulations in the region where you are implementing this technology, and research existing industry guidance to learn about policy guidelines and common fairness issues. Read about fairness in machine learning, including ways to mitigate bias in training datasets, evaluate your custom models for disparities in performance, and other considerations as you use your custom model.

We encourage customers to keep fairness, interpretability, and privacy and security best practices in mind when implementing Identity Proofing. To learn more about how to implement responsible AI, read Google's recommendations for Responsible AI practices.

Refer to the blog post Automate identity document processing with Document AI] for more information on use cases and a sample application code repository.