OCR PDF API
Make scanned PDFs searchable with OCR in 100+ languages — deskew, rotate, clean, and output PDF/A if you like.
Environment
| Environment | Base URL |
|---|---|
| Production | https://api.sadafiles.com/v1 |
There is one environment and no sandbox, so every request here is real and counts toward your limits.
Request
POSThttps://api.sadafiles.com/v1/tools/ocr-pdf
- Content type
- multipart/form-data
- Auth
- X-API-Key with tools:run
- Accepts
- Files per request
- 1 to 500
- Largest file
- 2 GB
- Job timeout
- 300 s
- Rate weight
- 2 units
- Web tool
- Use OCR PDF without code
Parameters
Send these as a JSON string in the options form field. Anything else is rejected.
| Name | Type | Default | Allowed |
|---|---|---|---|
languages | array | eng | — |
deskew | boolean | true | — |
rotate | boolean | true | — |
clean | boolean | false | — |
pdfa | boolean | false | — |
skipTextSkip pages that already have text | boolean | true | — |
Response
| Status | Body | When |
|---|---|---|
201 | pdf, zip | The run finished within 60 seconds |
202 | Job summary JSON | Still running; poll the job |
4xx, 5xx | Error JSON | See error codes |
A file response carries X-Sada-Job and X-Sada-Expires-At. For long work, post to /v1/tools/ocr-pdf/jobs instead. See jobs and webhooks.
Also runs on the device
In the web tool and apps, OCR PDF can run locally with no upload. The API always processes on the server.
Ready settings
These settings power the matching pages on the site, and work the same in the API.
| Use | options |
|---|---|
| Hindi OCR PDF | {"languages":["hin","eng"]} |
| Telugu OCR PDF | {"languages":["tel","eng"]} |
| Tamil OCR PDF | {"languages":["tam","eng"]} |
| Kannada OCR PDF | {"languages":["kan","eng"]} |
| Malayalam OCR PDF | {"languages":["mal","eng"]} |
Try it yourself!
Pick a file, set the options and run a real request with your key. The code follows your changes.