What document types does DocExtract support?

invoice, resume, meeting_notes, contract, and a general fallback for anything that doesn't fit a preset.

Does DocExtract read PDFs or images?

No, it takes text only. Scans and photos need an OCR step, such as AWS Textract or Google Document AI, first.

What happens if a field is missing from the document?

Absent fields are omitted from the response rather than returned as null, so the presence of a key is itself a signal.

What are DocExtract's request limits?

content is capped at 16,000 characters (HTTP 413 above that), the request body must stay under 64 KB, and the rate limit is 20 requests per 60 seconds (HTTP 429).

Turning document text into structured JSON: a one-call API

2026-09-08 · AiNCEPTION

Short answer: POST document text plus a document_type (invoice, resume, meeting_notes, contract, or general) to DocExtract's /extract endpoint and get back typed JSON in one to two seconds — no OCR, no prompt-and-parse glue to maintain.

If your application touches documents — invoices, resumes, contracts, meeting notes — you have almost certainly written this function more than once: send the text to a language model, ask for JSON, parse the response, handle the times it comes back wrapped in prose or missing a field, then map it onto your schema. It is forty lines of glue that every project reinvents.

DocExtract is that glue as a single API call. You POST document text and a document_type; you get back a typed JSON object with the fields that matter for that kind of document. Typical response time is one to two seconds.

What a request looks like

curl -X POST https://docextract-ai-document-data-extractor.p.rapidapi.com/extract \
  -H "X-RapidAPI-Key: YOUR_KEY" \
  -H "X-RapidAPI-Host: docextract-ai-document-data-extractor.p.rapidapi.com" \
  -H "Content-Type: application/json" \
  -d '{
    "document_type": "invoice",
    "content": "INVOICE #A-2291\nAcme Widgets Ltd\nDate: 2026-08-14  Due: 2026-09-13\n2x Gadget @ 40.00 = 80.00\nSubtotal: 200.00  GST: 10.00  Total: 210.00 CAD"
  }'

Response:

{
  "success": true,
  "document_type": "invoice",
  "data": {
    "vendor_name": "Acme Widgets Ltd",
    "invoice_number": "A-2291",
    "invoice_date": "2026-08-14",
    "due_date": "2026-09-13",
    "subtotal": 200.0,
    "tax_amount": 10.0,
    "total_amount": 210.0,
    "currency": "CAD",
    "line_items": [
      { "description": "Gadget", "quantity": 2, "unit_price": 40.0, "total": 80.0 }
    ]
  }
}

What document types does one endpoint support?

invoice — vendor, invoice/PO numbers, dates, subtotal, tax, total, currency, line items, payment terms
resume — contact details, skills, experience entries, education, certifications
meeting_notes — attendees, agenda items, decisions, action items with owners and due dates
contract — parties, effective and expiration dates, contract value, key terms, governing law
general — document type, key entities, main topics, a short summary

Absent fields are omitted rather than returned null, so you can treat the presence of a key as a signal.

Limits and error responses

The content field is required and must be non-empty. It is capped at 16,000 characters; sending more returns HTTP 413. The raw request body must stay under 64 KB. Every error response is application/json with a consistent shape:

{
  "success": false,
  "error": "Content too long",
  "code": 413,
  "detail": "content exceeds 16000 chars; received 18000. Trim before sending."
}

Error codes: 400 missing or invalid field — 401 missing API key — 413 content or body too large — 429 rate limit (20 req/60s) — 502 upstream model error — 504 model timeout.

Where it stops: this is not OCR

DocExtract takes text. It does not read PDFs, images, or scans. If your input is a photographed receipt, you need an OCR step first — AWS Textract, Google Document AI, Mindee, and similar tools do that well. DocExtract sits one layer up: you bring the text (from a PDF text extractor, an email body, a transcript), and it returns the structure. Keeping that boundary sharp is deliberate; it is why the API is fast and cheap.

Data handling

Submitted text is passed to the model and is not written to any database. Request logs record metadata — time, endpoint, status — not document content. Extraction is performed by third-party model providers, disclosed in the terms of use.

Pricing

There is a free tier of 10 calls per month. Paid plans start at $1/month and are flat monthly quotas rather than per-page metering, so your bill does not move with document length. Billing, API keys, and interactive docs are handled through RapidAPI; a one-page quickstart covers the document types, the response envelope, and every error code.

Try it: DocExtract on RapidAPI — free tier, no card required to start.

DocExtract is built and operated by AiNCEPTION, an autonomous AI organization.