Image to Text
A SMALL TOOL. A BIG PART OF YOUR WORKFLOW.

Image to text.
Copy. Edit. Keep building.

Turn screenshots, photos, and scans into editable text. Use free OCR in your browser, check the result beside your image, then copy or download TXT, Markdown, or JSON. Building an app or agent? Explore the API, CLI, and MCP documentation.

PRIVATE BY DEFAULT
01YOUR IMAGE

Drop your images here

or choose a file from your device

JPG, PNG, WebP · 10 MB each · up to 25 megapixels

No image handy? Try one:
02EXTRACTED TEXT Ready when you are
T

Your next copy & paste starts here.

Extract text, check it against the image, then copy or download.

.txt.md.json

Runs in your browser. Your images stay on your device.

No signup. No watermark. Yours to keep.Convert images in bulk →
BUILT FOR YOUR WORKFLOW
JSONREST APICLIMCPM↓ Markdown

LESS BUSYWORK. MORE BUILDING.

Goodbye, retyping.

For the little things that slow you down, and the big things you’re building.

01

Screenshots → editable text

Copy error messages, snippets, and text you cannot select.

02

Documents → searchable knowledge

Bring printed pages and scans into notes, search, and RAG workflows.

03

Receipts → reusable content

Extract receipt text, verify the details, and send it into your workflow.

BUILT TO FIT YOUR STACK

From a quick fix to an API call.

Bring text into your app, knowledge base, or agent. Same engine. Same credits. A workflow you control.

  • Async jobs & batch uploads
  • Idempotent requests
  • CLI & MCP included
  • OpenAPI 3.1 specification
Read the integration guide
extract.sh
# Pixels in. Text out.

curl https://imagetotext.dev/api/v1/convert \
  -H "Authorization: Bearer $IMAGETOTEXT_API_KEY" \
  -H "Idempotency-Key: example-conversion-0001" \
  -F "file=@screenshot.png;type=image/png" \
  -F "keepImages=false"
202 Accepted{ "id": "job_…", "status": "pending" }Illustrative submission response, not a live conversion result

How to convert an image to text

01 / Add your image

Drop a JPG, PNG, or WebP, browse your files, or paste a screenshot with Ctrl+V or ⌘V.

02 / Make it readable

Choose a language and run free OCR on your device. Use AI OCR for complex layouts and handwriting; review uncertain details.

03 / Take it anywhere

Edit, copy, and download your text. Connect the API, CLI, or MCP client when the job becomes a workflow.

Make the text work for you.

Get better image to text results

Start with a clear screenshot or a well-lit photo. Before uploading, straighten rotated pages and crop away unrelated content. Choose the language that matches the text, especially for documents containing more than one writing system. Small characters, shadows, compression, and decorative fonts can reduce recognition quality. After conversion, compare the result with the original image and check names, dates, amounts, and code symbols before using the text elsewhere.

Choose the right OCR mode

Free OCR runs in your browser and works best with clear printed text. Your image stays on your device while the browser downloads the recognition engine and language files. For handwriting or more complex layouts, consider AI OCR when it is available for your account. AI processing sends the selected image to a cloud provider and uses image credits. Choose the mode according to the document’s layout, privacy requirements, and the output you need.

Move the text into your workflow

Copy the result for a quick reply, download TXT for plain notes, or use Markdown in documentation and knowledge bases. JSON provides a machine-readable export for a later processing step. Check receipt amounts and table reading order before importing results into another system. Developers can explore authentication, asynchronous jobs, batch submissions, and plan requirements, while individual users can continue converting directly on this page.

Read the API documentation

A real image. An actual OCR result.

This original sample document was processed with the free browser OCR engine. Compare the image with its saved output, then try the same document in the converter.

Original sample · PNG · 1100 × 820
Sample printed document headed Field Notes with short sentences about turning screenshots into editable text

Saved free OCR output · English

| FIELD NOTES
Alittle clarity goes a long way.
Make room for your next idea.
Turn screenshots into editable text.
Keep your notes readable and searchable.
Copy what matters. Build what comes next.
Image to Text / Sample document

The saved result includes recognition imperfections: “A little” was read as “Alittle”. Review important text against the original.

Try this document in the converter

From an image to your next step.

Image to Markdown

Extract text, export a .md file, and learn when AI can help recover headings and tables.

Explore this workflow →

Screenshot to text

Paste a screenshot, choose its language, and copy editable text without creating an account.

Explore this workflow →

See the actual OCR results

Compare original images with unedited local and AI output, including mistakes and test conditions.

Explore this workflow →

One screenshot or your next product.

Free on your device. Simple image credits when you need AI. The same plans across website, API, and agents.

Free

$0

For the text you need right now.

  • Private, browser-based OCR
  • No account needed for local mode
  • Edit, copy, TXT / Markdown / JSON
  • 2 one-time AI image credits after signup

Basic

$9/mo

For everyday AI OCR on the website.

  • 100 AI images / month
  • AI OCR for photos, scans & handwriting
  • Preserve tables, code & headings
  • Website only · No API, CLI or MCP
  • Cancel any time

Personal

$19/mo

For power users, scripts, and AI agents.

  • 300 AI images / month
  • AI OCR for photos, scans & handwriting
  • Preserve tables, code & headings
  • API access and job history
  • Batch workspace, up to 10 images
  • CLI and MCP for scripts and AI agents
  • Cancel any time

Developer

$49/mo

For image-to-text in your application.

  • 2,000 AI images / month
  • AI OCR for photos, scans & handwriting
  • Preserve tables, code & headings
  • API access and job history
  • Batch workspace, up to 10 images
  • CLI and MCP for scripts and AI agents
  • Cancel any time

Team

$199/mo

For teams processing documents every month.

  • 10,000 AI images / month
  • AI OCR for photos, scans & handwriting
  • Preserve tables, code & headings
  • API access and job history
  • Batch workspace, up to 10 images
  • CLI and MCP for scripts and AI agents
  • Cancel any time

Enterprise

Custom

For volume, procurement, and custom workflows.

  • Custom image allowance by agreement
  • AI OCR for photos, scans & handwriting
  • Preserve tables, code & headings
  • API access and job history
  • Batch workspace, up to 10 images
  • CLI and MCP for scripts and AI agents
  • Discuss deployment & support requirements

Basic is website-only. Personal, Developer, Team, and Enterprise include API, CLI, MCP, and batch submissions. One AI image uses one credit; monthly credits expire after 31 days. Failed jobs restore credits. Enterprise requirements are scoped individually.

WHEN THE WORK GETS BIGGER

Image to text. At your scale.

Have a larger document workflow? Talk to us about volume, integration, procurement, and deployment requirements. We will scope the right setup together.

Explore Enterprise
  • Custom image volume
  • Integration & onboarding
  • Deployment requirements

Good questions. Clear answers.

Is the image to text converter free?

Yes. Free OCR runs in your browser without an account, watermark, or daily quota. AI OCR uses credits; signup includes 2 one-time trial images.

Which images and languages can I convert?

JPG, PNG, and WebP up to 10 MB and 25 megapixels. Free OCR offers English, simplified and traditional Chinese, Japanese, Korean, French, German, Spanish, Portuguese, Arabic, and Hindi. Select the language before converting.

Can I extract text from handwriting and tables?

AI OCR is designed for more complex layouts, handwriting, tables, and code. Free OCR works best on clear printed text. Recognition depends on image quality; always review important details against the original.

Are my images private?

In free mode, image bytes stay on your device. The browser downloads the OCR engine and language packs. AI mode uploads the selected image to Replicate; private results are saved to your account. The app does not retain original AI uploads. Provider retention policies also apply.

Can my app or AI agent use the image to text API?

Yes. Personal, Developer, Team, and Enterprise include API keys, async jobs, batch requests, a downloadable CLI, and a stdio MCP server. OpenAPI and llms.txt describe the same service. Basic and trial credits are website-only.

Does image to Markdown preserve tables and code?

Free OCR exports recognized text in a .md file without rebuilding document structure. AI OCR can infer headings, tables, and code formatting, but results vary. Review column order, numbers, and indentation before reuse. Our public OCR examples show unedited output from both modes.

How can I judge OCR accuracy before using it?

Try a representative image and compare the full result against the original. Check names, amounts, and technical symbols manually. We publish three synthetic source images with raw local and AI results and test conditions. These examples are evidence of those runs, not a universal accuracy percentage.