DocuStruct - PDF & Image to Structured JSON

DocuStruct - PDF & Image to Structured JSON Extraction API

PDF Extraction and Image to Structured JSON API for Document AI workflows. Extract factual B2B data from PDFs, invoices and technical images with schema-validated JSON output.

Search intent: PDF extraction · image to JSON · document AI · OCR · structured data

A real file in. Strict JSON out.

The playground sends the selected file to the protected server pipeline. The server validates its binary signature before storage and asks the multimodal model to return only the requested JSON shape.

Select a document and press Extract.

No result is fabricated before the server completes a request.

Use a product-scoped key outside RapidAPI.

Immediate API access after a successful subscription. The entitlement includes 1,000 processed pages/month and is revoked when the associated subscription ends.

curl --request POST \
  --url https://prodapi-gqe9kzue.manus.space/api/v1/docustruct/extract \
  --header 'X-API-Key: YOUR_DIRECT_KEY' \
  --header 'Content-Type: application/json'

RapidAPI-ready endpoint contracts.

Every endpoint documents a testable request shape or an explicit public-source requirement, plus a response example, so a developer can inspect the contract before running it in RapidAPI.

curl --request POST \
  --url https://YOUR_RAPIDAPI_HOST/api/v1/docustruct/extract \
  --header 'x-rapidapi-key: YOUR_KEY' \
  --header 'x-rapidapi-host: YOUR_RAPIDAPI_HOST' \
  --header 'Idempotency-Key: invoice-demo-001' \
  --header 'Content-Type: application/json' \
  --data '{"filename":"invoice.png","mimeType":"image/png","base64":"BASE64_FILE","schema":"{\"vendor\":\"string\",\"total\":\"number\"}"}'

{"id":"ds_...","status":"completed","pageCount":1,"result":{"vendor":"ACME","total":149.5},"model":"gemini-3-flash-preview"}

Commercial and integration questions.