OCR Clipboard

Every screenshot, searchable text.

KopyFeed automatically OCRs every image you copy in your browser. Receipts, code screenshots, whiteboard photos, business cards — every word, number, and symbol becomes searchable text. No separate OCR tool. No manual upload.

AI Vision

Sensitive content auto-blocked

100+ languages

A dashboard screenshot being processed by KopyFeed's OCR

Screenshot captured·figma.com

OCR Transcript

Active Users · 12,500

Churn Rate · 2.1%

MRR · $85,000

User Growth · Monthly Revenue · Sales Pipeline

Server Health · Real-time

Searchable across your feed

2 credits / image

How it works

Three steps. No new habits to learn. Every screenshot you take in your browser becomes searchable text automatically.

Capture an image

Press Alt+A for a full-viewport screenshot, Alt+S for region snip, or just copy any image with Ctrl+C. KopyFeed auto-saves it to your feed.

AI extracts the text

Google Gemini vision reads every word, number, and symbol visible in the image — up to 50,000 characters per image. No manual upload, no separate tool.

Search by content

The extracted text becomes part of your clip. Find a screenshot from weeks ago by typing a phrase that appears in it — receipts by total, code by function name.

What people actually OCR

Every screenshot you take has text in it. KopyFeed reads them all so you don't have to remember what each image was.

Receipts & Invoices

Screenshot a receipt today, find it by the total amount or vendor name months later.

Code Screenshots

Capture code from tutorials, Stack Overflow images, or paywalled docs. Search by function or library name.

Whiteboard Photos

Snap a meeting whiteboard, KopyFeed reads the marker scribbles. Find any session by what was discussed.

Business Cards

Photograph a business card, KopyFeed extracts the name, title, email, and phone number — searchable.

Foreign Language Images

Captures verbatim text in any of 100+ languages. Pair with KopyFeed summaries for context.

PDFs & Paywalled Content

Some PDFs disable copy. Screenshot the page — KopyFeed extracts the text into a searchable clip.

Charts & Dashboards

Capture analytics dashboards, financial charts, or product mockups. Axis labels, legends, and values become text.

Memes with Text

Keep your reaction-image stash organized. Find the right meme by the caption you remember.

Built for accuracy. Built for privacy.

Real OCR with a real model. Not a regex over alt-text. Not a half-baked extension feature.

AI vision model

Powered by Google Gemini 2.5 Flash — the same vision model that beats Tesseract on handwriting and matches Google Cloud Vision on print accuracy.

50,000 character transcript cap

Each image returns a verbatim transcript of every word, number, and symbol visible — up to 50,000 characters. Long documents get the full body, not a snippet.

Private by default

Images use the paid Gemini API tier — Google does not use your prompts or images to train their models. Sensitive content (API keys, credit cards, SSNs) is detected on-device before upload and never sent to AI.

2 credits per image

Cheap by design. Advanced ($6.99/mo) gets 1,500 credits/month — roughly 750 OCR'd images. Pro ($11.99/mo) gets 6,000 — about 3,000 images.

Read the full privacy policy on AI processing

Frequently asked questions

Everything about KopyFeed's OCR.

KopyFeed's OCR is powered by Google Gemini 2.5 Flash vision, which handles 100+ languages — including English, Spanish, French, German, Italian, Portuguese, Japanese, Korean, Chinese (simplified and traditional), Arabic, Hindi, Russian, and most European scripts. Mixed-language images (e.g., a screenshot with English text plus a Japanese URL) work in one pass.

On clean printed text, accuracy is essentially indistinguishable from the original — Gemini vision matches Google Cloud Vision on typical screenshots. On handwriting, accuracy is meaningfully better than older OCR engines like Tesseract but degrades with very loose cursive or low-contrast marker writing. For best results: high resolution, good contrast, no extreme rotation.

Yes. Whiteboard markers, notebook pages, post-it notes — Gemini reads them. Print handwriting performs best; cursive and slanted scripts work but with occasional substitutions. If a whiteboard photo matters, take a straight-on shot rather than an angled one.

Yes. Images are sent to Google's paid Gemini API tier — under Google's terms, paid-tier prompts and responses are not used to train Google's models. KopyFeed never resells or analyzes your images beyond providing your own search and AI features. Sensitive content (API keys, credit card numbers, AWS keys, SSNs) is detected on your device before upload and is never sent to AI. Free-tier users don't use cloud AI at all.

The OCR transcript is capped at 50,000 characters per image — enough for a full page of dense text. Images larger than 5MB are automatically JPEG-compressed to 80% quality before upload to stay within Gemini's limits without meaningful accuracy loss. Image thumbnails are 600px wide for fast feed scrolling.

Yes for both, with caveats. Code transcription is excellent — variable names, syntax, indentation all preserved. Math is read out (e.g., "x squared plus 2x plus 1") but the visible characters are extracted verbatim; full LaTeX reconstruction of complex equations is imperfect. For a code snippet from a tutorial, you'll get something you can copy and run.

No — cloud AI features (including OCR) require Advanced ($6.99/month) or Pro ($11.99/month). The free tier includes algorithmic categorization that runs on your device in under 10ms, but it doesn't include vision-based OCR. Free-tier users can still capture images; they just won't have the extracted text or AI description.

Standalone OCR (ABBYY, Tesseract, OCR.space) requires you to open a separate tool and upload each image manually — fine for occasional use, painful daily. Mac Live Text and Windows PowerToys Text Extractor work on your active screen but are OS-locked and don't store the extracted text searchably. KopyFeed runs in any Chromium browser on any OS, OCRs every image automatically as you copy it, and keeps the text searchable across your entire clipboard history.

Related features

More ways KopyFeed turns your clipboard into a searchable knowledge feed.

Stop losing screenshots. Search them instead.

KopyFeed launches soon. Join the waitlist — be first to OCR your clipboard.