Specialized
Specialized Forms Models
Download OpenAPI specification:Download
JSON response
Upload a social security card to extract Social Security Number data. Receive JSON response data.
File upload options
We support requests with Content-Type application/json
or multipart/form-data
. Note that application/json
requests are preferred as multipart/form-data
requests do not support all flags and fields.
[application/json]
inputURL link to file[application/json]
base64 base64 encoded file data[multipart/form-data]
file upload local file from browser
header Parameters
apiVersion required | string (apiVersion) non-empty YYYY-MM-DD Default: 2025-02-27 API Version |
orgId required | string (orgId) non-empty ^[a-zA-Z0-9\-_]+$ Organization ID |
authKey required | string (authKey) non-empty ^[a-zA-Z0-9\-_]+$ Authentication key |
Request Body schema: required
inputUrl required | string URL for the file to extract data from: Must be a PDF, JPEG, PNG, or TIFF. |
fileId | string Default: "UUID" Examples: "form1.pdf" Custom ID for document. If not present, will default to a random UUID. |
metadata | object Custom JSON to be included in the returned response. |
webhook | string Webhook URL to send status updates and JSON response to. |
webhookSendFull | boolean Default: false When set to |
forceBase64 | boolean Default: false Setting this parameter to true converts files to Base64 format before sending them on to the model. |
forceOCR | boolean Default: false Set this parameter to true if the input file is a fillable pdf. |
webhookHeaders | object Request headers to include in the POST request to the webhook. |
Responses
Request samples
- Payload
{- "inputUrl": "string",
- "fileId": "form1.pdf",
- "metadata": { },
- "webhook": "string",
- "webhookSendFull": false,
- "forceBase64": false,
- "forceOCR": false,
- "webhookHeaders": { }
}
Response samples
- 200
{- "documentId": "string",
- "metadata": { },
- "ssnData": {
- "ssnNumber": {
- "index": "string"
}
}, - "status": "string",
- "startTime": 0,
- "endTime": 0,
- "id": "string",
- "pages": 0,
- "modelType": "string"
}
Zip file response
Upload a social security card to extract SSN data and receive response data in a ZIP file.
File upload options
We only support requests with Content-Type application/json
at this endpoint.
[application/json]
inputURL link to file[application/json]
base64 base64 encoded file data
The default name for the file is an epoch timestamp if fileId
field is not included in the request.
header Parameters
apiVersion required | string (apiVersion) non-empty YYYY-MM-DD Default: 2025-02-27 API Version |
orgId required | string (orgId) non-empty ^[a-zA-Z0-9\-_]+$ Organization ID |
authKey required | string (authKey) non-empty ^[a-zA-Z0-9\-_]+$ Authentication key |
Request Body schema: required
inputUrl required | string URL for the file to extract data from: Must be a PDF, JPEG, PNG, or TIFF. |
outputUrl | string URL where resulting zip file should be sent. Must be open to PUT requests. Required if submitting an async request. |
outputUrlHeaders | object Request headers to include in the PUT request to the output URL. |
fileId | string Default: "UUID" Examples: "form1.pdf" Custom ID for document. If not present, will default to a random UUID. |
forceBase64 | boolean Default: false Setting this parameter to true converts files to Base64 format before sending them on to the model. |
forceOCR | boolean Default: false Set this parameter to true if the input file is a fillable pdf. |
statusId | string WARNING: Deprecating in favor of 'fileId'. Custom ID for document. If not present, will default to a random UUID. |
metadata | object Custom JSON to be included in the returned response. |
webhook | string Webhook URL to send status updates and JSON response to. |
webhookSendFull | boolean Default: false When set to |
webhookHeaders | object Request headers to include in the POST request to the webhook. |
Responses
Request samples
- Payload
{- "inputUrl": "string",
- "outputUrl": "string",
- "outputUrlHeaders": { },
- "fileId": "form1.pdf",
- "forceBase64": false,
- "forceOCR": false,
- "statusId": "string",
- "metadata": { },
- "webhook": "string",
- "webhookSendFull": false,
- "webhookHeaders": { }
}
JSON response
Upload an invoice for data extraction and receive JSON response data.
File upload options
We support requests with Content-Type application/json
or multipart/form-data
. Note that application/json
requests are preferred as multipart/form-data
requests do not support all flags and fields.
[application/json]
inputURL link to file[application/json]
base64 base64 encoded file data[multipart/form-data]
file upload local file from browser
header Parameters
apiVersion required | string (apiVersion) non-empty YYYY-MM-DD Default: 2025-02-27 API Version |
orgId required | string (orgId) non-empty ^[a-zA-Z0-9\-_]+$ Organization ID |
authKey required | string (authKey) non-empty ^[a-zA-Z0-9\-_]+$ Authentication key |
Request Body schema: required
inputUrl required | string URL for the file to extract data from: Must be a PDF, JPEG, PNG, or TIFF. |
fileId | string Default: "UUID" Examples: "form1.pdf" Custom ID for document. If not present, will default to a random UUID. |
metadata | object Custom JSON to be included in the returned response. |
webhook | string Webhook URL to send status updates and JSON response to. |
webhookSendFull | boolean Default: false When set to |
forceBase64 | boolean Default: false Setting this parameter to true converts files to Base64 format before sending them on to the model. |
forceOCR | boolean Default: false Set this parameter to true if the input file is a fillable pdf. |
webhookHeaders | object Request headers to include in the POST request to the webhook. |
Responses
Request samples
- Payload
{- "inputUrl": "string",
- "fileId": "form1.pdf",
- "metadata": { },
- "webhook": "string",
- "webhookSendFull": false,
- "forceBase64": false,
- "forceOCR": false,
- "webhookHeaders": { }
}
Response samples
- 200
{- "documentId": "string",
- "endTime": 0,
- "id": "string",
- "invoiceData": {
- "amountDue": "string",
- "customerAddress": "string",
- "customerAddressRecipient": "string",
- "customerName": "string",
- "dueDate": "string",
- "invoiceDate": "string",
- "invoiceId": "string",
- "invoiceTotal": "string",
- "items": [
- {
- "amount": "string",
- "description": "string",
- "quantity": "string",
- "unit_price": "string",
- "product_code": "string",
- "rawText": "string"
}
], - "keyValuePairs": [
- {
- "confidence": 0,
- "key": {
- "boundingRegions": [
- {
- "boundingBox": [
- 0
], - "pageNumber": 0
}
], - "content": "string",
- "spans": [
- {
- "offset": 0,
- "length": 0
}
]
}, - "value": {
- "boundingRegions": [
- {
- "boundingBox": [
- 0
], - "pageNumber": 0
}
], - "content": "string",
- "spans": [
- {
- "offset": 0,
- "length": 0
}
]
}
}
], - "pages": 0,
- "rawText": "string",
- "status": "SUCCESS",
- "subTotal": "string",
- "totalTax": "string",
- "vendorAddress": "string",
- "vendorAddressRecipient": "string",
- "vendorName": "string"
}, - "pages": 0,
- "startTime": 0,
- "status": "SUCCESS",
- "metadata": { }
}
Zip file response
Upload an invoice for data extraction and receive response data in a ZIP file.
File upload options
We only support requests with Content-Type application/json
at this endpoint.
[application/json]
inputURL link to file[application/json]
base64 base64 encoded file data
The default name for the file is an epoch timestamp if fileId
field is not included in the request.
header Parameters
apiVersion required | string (apiVersion) non-empty YYYY-MM-DD Default: 2025-02-27 API Version |
orgId required | string (orgId) non-empty ^[a-zA-Z0-9\-_]+$ Organization ID |
authKey required | string (authKey) non-empty ^[a-zA-Z0-9\-_]+$ Authentication key |
Request Body schema: required
inputUrl required | string URL for the file to extract data from: Must be a PDF, JPEG, PNG, or TIFF. |
outputUrl | string URL where resulting zip file should be sent. Must be open to PUT requests. Required if submitting an async request. |
outputUrlHeaders | object Request headers to include in the PUT request to the output URL. |
fileId | string Default: "UUID" Examples: "form1.pdf" Custom ID for document. If not present, will default to a random UUID. |
forceBase64 | boolean Default: false Setting this parameter to true converts files to Base64 format before sending them on to the model. |
forceOCR | boolean Default: false Set this parameter to true if the input file is a fillable pdf. |
statusId | string WARNING: Deprecating in favor of 'fileId'. Custom ID for document. If not present, will default to a random UUID. |
metadata | object Custom JSON to be included in the returned response. |
webhook | string Webhook URL to send status updates and JSON response to. |
webhookSendFull | boolean Default: false When set to |
webhookHeaders | object Request headers to include in the POST request to the webhook. |
Responses
Request samples
- Payload
{- "inputUrl": "string",
- "outputUrl": "string",
- "outputUrlHeaders": { },
- "fileId": "form1.pdf",
- "forceBase64": false,
- "forceOCR": false,
- "statusId": "string",
- "metadata": { },
- "webhook": "string",
- "webhookSendFull": false,
- "webhookHeaders": { }
}
Extract, JSON response
Make a request to extract PII data and receive JSON response data.
File upload options
We support requests with Content-Type application/json
or multipart/form-data
. Note that application/json
requests are preferred as multipart/form-data
requests do not support all flags and fields.
[application/json]
inputURL link to file[application/json]
base64 base64 encoded file data[multipart/form-data]
file upload local file from browser
header Parameters
apiVersion required | string (apiVersion) non-empty YYYY-MM-DD Default: 2025-02-27 API Version |
orgId required | string (orgId) non-empty ^[a-zA-Z0-9\-_]+$ Organization ID |
authKey required | string (authKey) non-empty ^[a-zA-Z0-9\-_]+$ Authentication key |
Request Body schema:
inputURL required | string non-empty https?://(?:www\\.)?[a-zA-Z0-9./]+ File to upload: Must be a PDF, JPEG, PNG, or TXT. Note that TIFF files are currently unsupported for PII. |
forceBase64 | boolean Default: false Setting this parameter to true converts files to Base64 format before sending them on to the model. |
pageQualityWeight | number <float> [ 0 .. 1 ] Default: 0.5 Hyperparameter to weigh word confidences against confidence that a given word is PII. |
redact | boolean Default: false Set to true to include the redacted file in the response as a base64-encoded PDF. |
Responses
Request samples
- Payload
{- "inputURL": "string",
- "forceBase64": false,
- "pageQualityWeight": 0.5,
- "redact": false
}
Response samples
- 200
{- "status": "SUCCESS",
- "items": [
- {
- "dimension": {
- "height": 0,
- "width": 0,
- "unit": "string"
}, - "regions": [
- {
- "polygon": [
- 0
], - "content": "string",
- "confidence": 1
}
]
}
], - "redactedFile": {
- "data": "string",
- "encoding": "string"
}
}
Redact, JSON response
Make a request to redact PII data from a document and receive JSON response data.
File upload options
We support requests with Content-Type application/json
or multipart/form-data
. Note that application/json
requests are preferred as multipart/form-data
requests do not support all flags and fields.
[application/json]
inputURL link to file[application/json]
base64 base64 encoded file data[multipart/form-data]
file upload local file from browser
header Parameters
apiVersion required | string (apiVersion) non-empty YYYY-MM-DD Default: 2025-02-27 API Version |
orgId required | string (orgId) non-empty ^[a-zA-Z0-9\-_]+$ Organization ID |
authKey required | string (authKey) non-empty ^[a-zA-Z0-9\-_]+$ Authentication key |
Request Body schema:
inputURL required | string non-empty https?://(?:www\\.)?[a-zA-Z0-9./]+ URL for the file to extract data from: Must be a PDF, JPEG, PNG, TIFF, or TXT. Note that TIFF files are currently unsupported for PII. |
forceBase64 | boolean Default: false Setting this parameter to true converts files to Base64 format before sending them on to the model. |
getPII | boolean Set to true to run the PII model first and redact any identified regions. |
pageQualityWeight | number <float> [ 0 .. 1 ] Default: 0.5 Hyperparameter to weigh word confidences against confidence that a given word is PII. |
keywords | Array of strings Labels of fields to redact. |
Array of objects (PageContent) List of page content to redact regions to redact. Can directly insert results from /pii response field "items" or custom regions and dimensions. |
Responses
Request samples
- Payload
{- "inputURL": "string",
- "forceBase64": false,
- "getPII": true,
- "pageQualityWeight": 0.5,
- "keywords": [
- "string"
], - "pages": [
- {
- "dimension": {
- "height": 0,
- "width": 0,
- "unit": "string"
}, - "regions": [
- {
- "polygon": [
- 0
], - "content": "string",
- "confidence": 1
}
]
}
]
}
Response samples
- 200
{- "status": "SUCCESS",
- "redactedFile": {
- "data": "string",
- "encoding": "string"
}
}