Built for developers and document pipelines

OCR API for PDFs, scans, and document pipelines

Turn PDFs and document images into structured JSON, Markdown, plain text, HTML, or DOCX. Upload once, poll an asynchronous job, and fetch a consistent result from one REST API.

Process 30 pages free Read API docs

30 pages free · no credit card · test in the browser before choosing API access

free test pages

$0.005

from Flex pricing

input file types

result formats

quickstart.sh

Request

curl -X POST https://sotaocr.com/v1/extract \
  -H "Authorization: Bearer YOUR_API_KEY" \
  -F "file=@document.pdf"

Result

{
  "id": "job_123",
  "status": "completed",
  "page_count": 5,
  "model_profile": "fast"
}

POST /v1/extract →GET /v1/jobs/{job_id}/result

A small, predictable API

From document to usable output in three calls

Keep OCR outside your application process. SotaOCR handles preprocessing and long-running jobs while your code keeps a simple asynchronous contract.

Upload a document

Send PDF or image bytes to POST /v1/extract with your Bearer API key.

Poll job status

Read progress from GET /v1/jobs/{job_id}; large files do not hold an HTTP request open.

Fetch the format you need

Return structured JSON, Markdown, text, standalone HTML, or a reconstructed DOCX.

Designed for real documents

One pipeline, several downstream uses

Use text for search, Markdown for LLM context, structured page data for extraction, or visual reconstruction when layout matters.

JSONMarkdownPlain textHTMLDOCX

Async by default

Process multi-page files without fragile request timeouts.

Multilingual documents

Use the same integration for English and multilingual document sets.

Document-aware output

Preserve pages, preprocessing metadata, and coordinates in JSON.

Integration-friendly

Bearer API keys, explicit statuses, and documented error codes.

Transparent pricing

Start with a free evaluation, then pay for real usage

The free 30-page test lets you validate output in the browser. Flex and subscriptions add API access for production integration.

Flex

$0.005/page

Buy only the pages you need

No monthly commitment

Starter

$8/month

2,000 pages included

For prototypes and small workloads

Popular

Growth

$29/month

8,000 pages included

For production document flows

Scale

$99/month

30,000 pages included

For higher-volume processing

Process 30 pages free

Clear operational boundaries

You should know how an OCR service fits into your system before sending production documents. The API uses account-scoped Bearer API keys, and API job artifacts are automatically pruned after 24 hours by default.

Before you integrate

Authentication uses Bearer API keys created in your account.
Supported uploads: PDF, PNG, JPG, JPEG, WEBP, BMP, TIF, and TIFF.
Polling and error behavior are documented with explicit HTTP statuses.
Page previews are available when you need the image coordinate space behind OCR boxes.
Talk to support before sending regulated or unusually sensitive document classes.

Test your own documents before you commit

Create an account, process 30 pages in the browser, inspect the output, and move to API access only when the result fits your pipeline.

Process 30 pages free

No credit card required for the evaluation.