* * * PENNYOCR × LLAMAINDEX * * *
PennyOCR with LlamaIndex.
LlamaIndex is only as good as the Documents you feed it — and scans feed it nothing without OCR. A PennyOCR-backed reader turns any PDF or image into markdown Documents with page metadata: pip install pennyocr llama-index, ten lines (below).
START FREE — 100 PAGES
$0.75 / 1,000 PAGES
1,333 PAGES PER DOLLAR · NO CREDIT CARD
01 / TRY IT
ONE ENDPOINT.
POST a file, get JSON back — the extracted text, per page and joined. PDF, PNG, JPEG, WebP or TIFF.
curl https://api.pennyocr.com/v1/ocr \
-H "Authorization: Bearer $PENNYOCR_API_KEY" \
-F "file=@scan.pdf"
# $0.75 per 1,000 pages, first 100 free02 / USE CASES
THE READER.
THE CODE
from pennyocr import PennyOCR · from llama_index.core import Document · docs = [Document(text=p.text, metadata={"page": p.page, "source": path}) for p in PennyOCR().ocr(path).page_results]
STRUCTURE-AWARE NODES
MarkdownNodeParser splits along the headings and tables PennyOCR preserves — nodes that mean something.
CITATION QUERIES
Page metadata flows into CitationQueryEngine, so answers cite page numbers you can check.
COMPARED TO LLAMAPARSE
LlamaParse is polished and deeper on layout modes; check its current per-page credits pricing against our flat $0.00075/page and run both on your documents.
03 / PRICE CHECK
HALF THE PRICE OF THE BIG CLOUDS.
Per 1,000 pages, public list prices, first tier.
PENNYOCR$0.75
AWS TEXTRACT$1.50
GOOGLE CLOUD VISION$1.50
AZURE DOC INTELLIGENCE$1.50
YOU KEEP50%
OFFICIAL READER PACKAGE?
On the roadmap; the ten-liner works today. Tell hello@pennyocr.com if it should ship sooner.
SCANNED + DIGITAL MIXED?
Yes — everything is read visually, so one reader covers the whole corpus with one output format.
WHAT DOES INGESTION COST?
$0.75 per 1,000 pages, prepaid, first 100/month free. A 50k-page knowledge base is $37.50, once.
ALSO ON THE MENU