Pricing

from $4.00 / 1,000 page ocr'ds

Document OCR | PDF to Markdown | Invoice, Table & Image to Text

Document OCR and PDF to Markdown. Extract text, tables and structured data from PDFs and images with state-of-the-art PaddleOCR-VL. Handles invoices, receipts, forms and scanned documents in 100+ languages, returning LLM-ready Markdown plus layout blocks for RAG. Pay per page.

Pricing

from $4.00 / 1,000 page ocr'ds

Rating

0.0

(0)

Developer

Scott Shinn

Actor stats

Bookmarked

Total users

Monthly active users

a day ago

Last modified

Document OCR API — PDF & Image to Markdown (PaddleOCR-VL)

Give this Actor a list of image or PDF URLs. Get back clean Markdown and structured layout blocks — tables kept as tables, headings kept as headings, reading order preserved — ready to drop into a RAG pipeline, a search index, or a data-extraction step. Runs on GPU with PaddleOCR-VL, across 100+ languages.

See it work

Input:

{ "imageUrls": ["https://templates.invoicehome.com/invoice-template-us-neat-750px.png"] }

Output (one dataset item — trimmed):

## INVOICE

East Repair Inc.
1912 Harvest Lane
New York, NY 12210

INVOICE #  US-001
INVOICE DATE  11/02/2019
DUE DATE  26/02/2019

…and the invoice's line-item table comes back as a real table, not flattened text:

QTY	DESCRIPTION	UNIT PRICE	AMOUNT
1	Front and rear brake cables	100.00	100.00
2	New set of pedal arms	15.00	30.00
3	Labor 3hrs	5.00	15.00

Every page also returns blocks — the layout regions with their type and bounding-box coordinates — if you want to post-process programmatically.

Input

Field	Type	Description
`imageUrls`	array of strings	Required. Publicly accessible image or PDF URLs. Each PDF is OCR'd page by page.
`maxPages`	integer	Optional. Cap pages per PDF (`0` = no cap).

Output

One dataset item per URL:

{
  "url": "...",
  "ok": true,
  "numPages": 1,
  "markdown": "## INVOICE\n\nEast Repair Inc. ...",
  "pages": [{ "markdown": "...", "blocks": [ /* type + bbox per region */ ] }]
}

A URL that fails comes back as { "url": "...", "ok": false, "error": "..." }, so one bad link never kills the run.

What it's good at

Tables and forms — invoices, receipts, financial statements come back with rows and columns intact.
Scanned documents and photos — printed text is the sweet spot; handwriting and degraded scans work with lower confidence.
Multi-column and math-heavy pages — research papers, books, reports keep their reading order and formulas.
100+ languages — Latin, Cyrillic, Arabic, Devanagari, CJK, Thai, and more.

The engine is PaddleOCR-VL-1.6, which sits at the top of the OmniDocBench document-parsing benchmark — ahead of general-purpose vision models many times its size.

How it compares

Tool	Output	Trade-off
This Actor	Markdown + layout blocks	Top OmniDocBench accuracy; pages run one at a time
Tesseract actors	Plain text	Faster, weaker on tables and complex layout
Google Vision / AWS Textract	Text + basic layout	Robust, but per-call pricing and less LLM-ready output

Pricing

Pay per page OCR'd — one image is one page, a 10-page PDF is ten. No subscription, no minimum.

Notes

URLs must be publicly reachable.
Pages run one at a time through a single warm GPU model, so large batches proceed at a steady rate.

OCR & Document Extractor – PDF & Image to Text, JSON, Word

lofomachines/ocr-document-extractor

Convert scanned PDFs and images into clean, structured text in bulk. Export to JSON, Markdown, DOCX, TXT or HTML with tables and layout preserved.

Lofomachines

PDF to Markdown Converter - AI-Powered with OCR & Tables

clearpath/pdf-to-markdown-api

Convert PDFs to clean Markdown with GPU-accelerated AI. Extracts tables, LaTeX formulas, and images from complex layouts. Supports OCR for scanned docs in 8 languages. Batch process hundreds of PDFs in parallel via URL, upload, or API.

ClearPath

PDF Intelligence

marielise.dev/pdf-intelligence

Stop fighting PDFs. Extract text, tables, and insights from any document, scanned or digital. Get RAG-ready chunks for LangChain & LlamaIndex. AI-powered summaries, classification, entity extraction. Use our API keys or bring your own (50% discount). From PDF chaos to clean data in minutes.

Marielise

Document Extractor API - AI-Powered PDF & Text Analysis

fresh_cliff/document-extractor-api

Extract text and data from PDF, Word, and image documents using AI-powered OCR. Convert documents to structured JSON, analyze content, and extract insights. No API keys required with mirror fallbacks.

Brennan Crawford

PDF Text Extractor — Markdown, OCR & Metadata

memo23/pdf-text-extractor

Turn any PDF URL into clean data: full text, LLM-ready Markdown, per-page content, and metadata. Reads scanned PDFs via OCR (23 languages), supports page ranges, and processes files in parallel. Export JSON, CSV, or Excel. Ideal for RAG pipelines, invoices, and research.

Muhamed Didovic

113

Markitdown Mcp Server

rector_labs/markitdown-mcp-server

Cloud-hosted MCP server converting 29+ document formats (PDF, DOCX, PPTX, images, audio) to AI-ready Markdown. Zero Python setup. Perfect for RAG pipelines and AI agents. Pay-per-use: $0.02/conversion. Built on Microsoft's Markitdown (82k+ ⭐).

RECTOR SOL

PDF to Markdown Converter - Extract & Format Text

ntriqpro/pdf-to-markdown

Convert PDF documents into clean, readable Markdown with the document structure preserved — headings, lists and paragraph flow. Process files in bulk and get output that drops straight into LLM pipelines, RAG systems, knowledge bases and documentation sites.

daehwan kim

PDF Text Extractor & Document Parser: DOCX, XLSX, OCR, Markdown

yabanana99/pdf-word-excel-to-markdown

Parse Word, PowerPoint, Excel and PDF documents into clean Markdown and structured JSON: tables extracted cell-by-cell, document metadata, and RAG-ready chunks with heading paths. Built-in OCR for scans and images.

Daniele Giovane

PDF OCR Extractor — scanned PDFs to text (Arabic + English)

amanatools/pdf-ocr-extractor

Turn scanned or image-only PDFs into structured text that keeps its reading order: per-block bounding boxes, per-block OCR confidence, column-aware ordering, and a spatial plain-text rendering. Arabic and English OCR built in. Pay per document plus per page.

Dos

PDF Text Extractor — Structured Tables, Markdown & Metadata

nuhemugames/pdf-text-table-extractor

Extract text from PDF URLs at scale: full text, per-page text, real structured tables (rows and columns as JSON, not text lines), document metadata and clean Markdown for LLM/RAG. The table extraction generic PDF text extractors don't have. No credentials needed.

kuon

PDF to Markdown & JSON Extractor for LLMs

f0rty7even/pdf-extractor

Turn any PDF URL into clean, LLM-ready Markdown and structured JSON. Extracts text + tables + document metadata for RAG, agents, and fine-tuning. No AGPL components.