Documents (OCR)

Transform any document into AI-ready structured content with our powerful document processing API.

The Documents API enables you to extract text, tables, and images from a wide variety of file formats including PDFs, Microsoft Office documents (Word, PowerPoint, Excel), images, and more. Using advanced OCR technology, even scanned documents and images with embedded text can be converted into clean, structured output formats like Markdown, HTML, or JSON. Perfect for building RAG pipelines, document digitization workflows, invoice processing, and any application that needs to understand document content.

Authentication

All requests require an API key passed as a Bearer token in the Authorization header:

Authorization: Bearer YOUR_API_KEY

Supported file formats

The Documents API accepts a wide range of document and image formats.

Output formats

Choose one or more output formats for your processed documents.

Highlights

• Wide format support: process PDFs, Office documents, images, and more in a single API.
• Table extraction: fast or accurate modes for table structure detection.
• Multiple output formats: get results in Markdown, JSON, HTML, or plain text.

API endpoint

Process documents and images using multipart/form-data.

POST https://api.greenpt.ai/v1/tools/documents/convert/file

Note: This endpoint uses multipart/form-data for file uploads.

Basic example

A simple request to convert a PDF to markdown.

curl -X POST https://api.greenpt.ai/v1/tools/documents/convert/file \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "files=@document.pdf"

import fs from 'node:fs';

const form = new FormData();
form.append(
  'files',
  new Blob([fs.readFileSync('document.pdf')]),
  'document.pdf',
);

const response = await fetch(
  'https://api.greenpt.ai/v1/tools/documents/convert/file',
  {
    method: 'POST',
    headers: {
      Authorization: 'Bearer YOUR_API_KEY',
    },
    body: form,
  },
);

const result = await response.json();
console.log(result);

import requests

url = "https://api.greenpt.ai/v1/tools/documents/convert/file"

headers = {
  "Authorization": "Bearer YOUR_API_KEY"
}

with open("document.pdf", "rb") as f:
  files = {"files": ("document.pdf", f)}
  response = requests.post(url, headers=headers, files=files)

print(response.json())

Advanced example

A more advanced request with OCR and table extraction options.

curl -X POST https://api.greenpt.ai/v1/tools/documents/convert/file \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "files=@document.pdf" \
  -F "to_formats=md" \
  -F "do_ocr=true" \
  -F "table_mode=accurate" \
  -F "do_table_structure=true" \
  -F "include_images=true"

import fs from 'node:fs';

const form = new FormData();
form.append(
  'files',
  new Blob([fs.readFileSync('document.pdf')]),
  'document.pdf',
);
form.append('to_formats', 'md');
form.append('do_ocr', 'true');
form.append('table_mode', 'accurate');
form.append('do_table_structure', 'true');
form.append('include_images', 'true');

const response = await fetch(
  'https://api.greenpt.ai/v1/tools/documents/convert/file',
  {
    method: 'POST',
    headers: {
      Authorization: 'Bearer YOUR_API_KEY',
    },
    body: form,
  },
);

const result = await response.json();
console.log(result);

import requests

url = "https://api.greenpt.ai/v1/tools/documents/convert/file"

headers = {
  "Authorization": "Bearer YOUR_API_KEY"
}

data = {
  "to_formats": "md",
  "do_ocr": "true",
  "table_mode": "accurate",
  "do_table_structure": "true",
  "include_images": "true"
}

with open("document.pdf", "rb") as f:
  files = {"files": ("document.pdf", f)}
  response = requests.post(url, headers=headers, files=files, data=data)

print(response.json())

Parameters

The most common parameters are below. Only two defaults are applied by the API: to_formats defaults to ["md"] and do_ocr defaults to "true". Any other parameter you omit is left to the underlying Docling engine, whose defaults are shown as "Docling default" for reference and may change with the engine version.

Advanced options

These are passed through to Docling when provided. Omit them to use the engine's defaults.

Response format

{
  "document": {
    "filename": "document.pdf",
    "md_content": "# Document Title\n\nExtracted content...",
    "json_content": {
      "schema_name": "DoclingDocument",
      "version": "1.8.0",
      "name": "document.pdf",
      "origin": {
        "mimetype": "application/pdf",
        "filename": "document.pdf"
      },
      "body": { },
      "texts": [ ],
      "tables": [ ],
      "pictures": [ ],
      "pages": { }
    },
    "html_content": "<html>...</html>",
    "text_content": "Plain text content..."
  },
  "status": "completed",
  "errors": [],
  "processing_time": 2.45,
  "timings": {
    "ocr": 1.2,
    "parsing": 0.8,
    "export": 0.45
  }
}

Use cases

• Document digitization: convert scanned documents and PDFs to searchable text.
• Data extraction: extract tables from financial reports and spreadsheets.
• Invoice processing: automate invoice and receipt data extraction.
• Academic research: process research papers with formulas and citations.
• Legacy migration: convert old document formats to modern standards.
• Accessibility: make image-based documents accessible with text extraction.
• Content indexing: prepare documents for search engine indexing.
• Translation prep: extract text for translation workflows.