Google Document AI

API

Google Cloud API for OCR, form parsing and invoice data extraction

Price: Free tier, then pay per use
Access: Google Cloud credentials

Docs

About

Google Cloud service that runs documents through processors: Enterprise OCR, Form Parser, Layout Parser, custom extractors and pretrained parsers for invoices, receipts and IDs. You create a processor per project and region before sending files; it suits teams already on Google Cloud.

What you can do with it

Digitize scanned documents into text with layout, handwriting and quality scores
Pull key-value pairs, tables and entities out of forms, invoices and receipts
Train a custom extractor or classifier for your own document types

Get started

Enable the Document AI API in a Google Cloud project
Create a processor, such as Enterprise Document OCR, in us or eu
Save the request body with base64 file content as request.json

Example

# request.json: {"rawDocument": {"mimeType": "application/pdf", "content": "<base64>"}}
curl -X POST \
  -H "Authorization: Bearer $(gcloud auth print-access-token)" \
  -H "Content-Type: application/json; charset=utf-8" \
  -d @request.json \
  "https://us-documentai.googleapis.com/v1/projects/$PROJECT_ID/locations/us/processors/$PROCESSOR_ID:process"

Details

Hosting: Hosted service
Available in: Worldwide
Official SDKs: C#, Go, Java, JavaScript/TypeScript, PHP, Python, Ruby
MCP server: None
Works with: Google Cloud