Extraction API fordiagram-heavy documents.
Flowcharts, schematics, figures — extracted as complete image crops, not text approximations. Plus reading-order text, tables, math, and reference linking across PDF, PPTX, and images.
200 pages free·Files up to 100 MB·Pay as you go·Complete figure crops, never clipped
Trusted by developers building with
JavaScript
Python
Node.js
PHP
.NET
Go
Built for the document workflows that actually pay.
One API turns PDFs, slides, images, and scans into structured JSON and Markdown. Drop it into RAG pipelines, automation workflows, technical-document tooling, or slide processing — same endpoint, same shape.
RAG pipelines
Clean Markdown chunks with tables, math, and figure references preserved — drop into your vector store.
Document automation
Structured JSON for invoice, form, receipt, and contract workflows. Webhooks + idempotency built in.
Technical documents
Diagrams, schematics, flowcharts extracted as complete crops — not lossy text descriptions.
PPTX → JSON
Slides, speaker notes, slide numbers, embedded charts — byte-exact from PowerPoint XML.
- PPTX
- PPT
- PNG
- JPEG
- WebP
- HEIC
- HEIF
- Scans
- JSON
- Markdown
Up and running in 60 seconds.
One install, one call. Pick polling, a webhook, or sync — the response shape is the same. Every knob is on the right, so you never need to leave the page to start.
# pip install ocrqueen
from ocrqueen import OCRQueen
client = OCRQueen() # reads OCRQUEEN_API_KEY from env
with open("document.pdf", "rb") as f:
job = client.extract.create(file=f)
# Polls with exponential backoff until the job is terminal.
final = client.jobs.wait(job)
print(final.markdown) # rendered Markdown
# final.document — full structured JSONTry it free
Run your hardest document through OCRQueen.
Drop a PDF, PPTX, or iPhone photo into the playground and get clean, structured JSON + Markdown back in seconds. First 200 pages free.