Image to Text OCR: Extract Text from Images [$2.40/1K 💰]

Inkscry: Image to Text OCR (Extract Text from Images)

Get the text out of screenshots, scans, photos and receipts, in 100+ languages

Inkscry reads the text in your images with the Tesseract OCR engine and returns it as clean text with a confidence score, plus every line (and optionally every word) with its bounding box. It cleans each image first (fixes phone rotation, removes transparency, boosts contrast, upscales small screenshots), which noticeably improves accuracy on real-world pictures. $2.40 per 1,000 images, no start fee.

▶ Try it now with a prefilled input · $0.0024 per image (100 images = $0.24) · images with no text are free · Apify's free plan includes $5 of credit a month

Built for: teams digitising scanned pages, receipts, invoices and forms · marketers and researchers pulling text from screenshots, ads and social images · developers who need an OCR API without running Tesseract · AI agents that must "read" an image before reasoning about it

Real output (real run on Apify, 10 Nov 2026: 3 links in 4.7 s at 2 GB, platform cost $0.0006; the photo with no text and the PDF link were not charged):

Image	Text (start)	Confidence	Words
Tesseract sample page (eng_bw.png, 1486×668)	Mild Splendour of the various-vested Night! / Mother of wildly-working visions! hail / I watch thy gliding, while with watery light…	93	64
Example.jpg (a photo, no text)	(no text found: not charged)	0	0

{"imageUrls": ["https://tesseract.projectnaptha.com/img/eng_bw.png"]}

What you get for each image

Field	What it is
imageUrl, status	The image and ok, no-text or error.
text	The text in reading order; paragraphs separated by a blank line.
confidence	Average OCR confidence, 0–100.
lines	Each line: text, confidence and bbox (x, y, width, height in original pixels).
wordBoxes	Each word with confidence and box (when includeWords is on).
words, characters	Size of the text found.
width, height, format, fileSizeBytes	About the image.
languages	The OCR languages used.
sha256, change	Image fingerprint; new or changed in monitor mode.
error	Why an image failed (404, not an image, too big).

Ready-made inputs

Job	Input
Screenshot to text	{"imageUrls": ["https://example.com/screenshot.png"]}
German and English scans	{"imageUrls": ["https://example.com/brief.jpg"], "languages": ["German", "English"]}
Japanese / Chinese	{"imageUrls": ["https://example.com/menu.jpg"], "languages": ["Japanese"]}
Receipts with word boxes	{"imageUrls": ["https://example.com/receipt.jpg"], "includeWords": true, "minConfidence": 60}
Signs, labels and memes	{"imageUrls": ["https://example.com/sign.jpg"], "layout": "sparse"}
New images only, daily	{"imageUrls": ["https://example.com/daily-chart.png"], "monitor": true}

Use cases

• Document digitisation: scanned letters, forms, book pages and faxes to searchable text.
• Receipts and invoices: read totals, dates and vendor names; word boxes let you pick fields by position.
• Screenshots and social images: pull quotes, prices and captions out of images for analysis.
• Accessibility and SEO: draft alt text and transcripts for image-heavy pages.
• AI agents: give an LLM the text inside an image in one tool call.

Combine with Picpail to collect every image on a website, then read them all with Inkscry.

Run it on a schedule

1. Save your input with "monitor": true as a task.
2. Schedules → Create: daily (cron 0 6 * * *).
3. Add an integration (email, Slack, webhook to Make/Zapier/n8n). Only new or changed images come back; unchanged ones are free.

Integrations

• API (curl):

  curl -X POST "https://api.apify.com/v2/acts/cybermax~image-to-text-ocr/run-sync-get-dataset-items?token=YOUR_TOKEN" \
    -H "Content-Type: application/json" -d '{"imageUrls": ["https://tesseract.projectnaptha.com/img/eng_bw.png"]}'
  

• Python: ApifyClient(token).actor("cybermax/image-to-text-ocr").call(run_input={"imageUrls": [...], "languages": ["English"]})
• Make, Zapier, n8n: Apify app or node, run cybermax/image-to-text-ocr, map text and confidence into Sheets, Airtable or your LLM step.
• AI agents (MCP): https://mcp.apify.com?actors=cybermax/image-to-text-ocr.

Why Inkscry

	Inkscry	Typical image OCR Actor on the Store
Image clean-up	Rotation, transparency, contrast and small-image upscaling built in	Often raw image only
Output	Text + confidence + line and word boxes in original pixels	Often text only
Languages	100+ (any Tesseract language), several at once	Often English or a short list
No-text images	Free	Usually charged
Monitor mode	Only new or changed images	Rare
Price	$2.40 per 1,000 images, no start fee	$2.50–$5 per 1,000 images, most with start fees (Store prices on 10 Nov 2026)

FAQ

How accurate is it? Printed text in clear images is usually read at 90%+ confidence. Handwriting, very stylised fonts and blurry photos are harder for any OCR engine; check confidence and use minConfidence to drop noise.

Can it read PDFs? PDFs with a text layer are faster and cheaper with Pageshuck: PDF to Text & Markdown. For scanned pages, export them as images and read them here.

What sizes work? Up to 25 MB per image by default (100 MB max) and 120 megapixels.

Is it legal? Inkscry reads images you link to. Use it on images you own or may process; it collects nothing else and stores nothing after your run.

Why not Google Vision or AWS Textract? They need a cloud account and API keys and bill per 1,000 pages at higher prices; Inkscry runs on your Apify account with no extra setup.

Also from CyberMax

• Pageshuck: PDF to Text & Markdown · PDF, Word and PowerPoint to text.
• Picpail: Bulk Image Downloader · every image from a list of pages.
• Pixkiln: Image Converter & Compressor · resize and convert images in bulk.

Found this useful? A quick rating helps others find it.

© 2026 CyberMax. All rights reserved.