Free document tool
Free Image to Markdown
Turn visible text in a scan, photo or screenshot into editable Markdown.
Upload one document image. The selected OCR provider reads its pixels and rebuilds headings, paragraphs, lists and simple tables in Markdown. Keep the image open beside the text and check names, punctuation, numbers and narrow table cells before you copy or download anything.
Content reviewed .
Copy and Markdown download are available from the Text viewer after conversion. See Free limits.
Anonymous originals stay in browser storage, not cloud file storage. Extraction may send document content to the selected AI provider.
Upload a document image. JPG, PNG, WebP, GIF, TIFF or AVIF image files are supported up to 150 MB.
28-day Free trial · 25 page-equivalents · 5 hosted pages/document
Sample images and reviewed results
Reference results checked against each fictional image and generated with our download exporter. Fresh AI extractions may differ; review your result before using it.
The prepared image is a document screenshot with a large title, two section headings, body text, a short checklist, and a compact three-column table. Because an image has no text layer, the workflow must read characters from pixels before it can decide where Markdown headings, paragraphs, lists, and table rows belong. The example includes slightly uneven contrast and a cropped page edge so the result can be checked against realistic OCR problems instead of an artificially perfect text export.
The reviewed result keeps the visible wording, restores the heading levels, turns the checklist into Markdown list items, and expresses the simple grid as a pipe table. We compare names, dates, punctuation, quantities, and table cells with the source image. OCR can still confuse similar characters or miss text hidden by blur, glare, compression, or cropping. The editable result therefore opens beside the image so those mistakes can be corrected before copying or downloading the Markdown.
Medieval manuscript page: Public-domain OCR sample
Sample image
German handwritten invoice: Public-domain OCR sample
Sample image
The output stays beside the evidence
Dynamite Docs keeps the source and structured result in one review workspace. Infer document fields and repeating rows, correct values against the page, inspect warnings and export the accepted result.
This is more than a plain OCR transcript. The result separates fields, line items, totals and notes so a person can check the structure before sending data to Excel, CSV, JSON, Google Sheets, Google Drive, a webhook or the API.
Connect an AI provider and get a key
Choose a hosted model or connect your own provider key for the extraction. Open AI & Processing, choose Connect Provider and follow the provider link to create a key. Several providers offer free developer allowances. Availability and limits come from the provider.
Paste the key once, verify it and select a model for the run. Provider keys are encrypted before storage. Find a provider and key or follow the setup guide.
What the editable Markdown contains
- Visible headings and paragraphs become editable Markdown after the selected OCR provider reads the image pixels.
- Lists and simple tables can retain Markdown structure instead of arriving as one unbroken block of OCR text.
- The source image remains available beside the Text viewer so characters, punctuation, and table cells can be checked before reuse.
- The output preserves visible wording rather than summarizing the page. Reviewers can change heading levels, list nesting or table syntax without treating an OCR guess as part of the source.
- Markdown captures semantic structure, not the original design. Fonts, colors, precise spacing, complex diagrams and decorative placement are not recreated in the text file.
From visible pixels to Markdown
An image contains pixels rather than selectable text. OCR must recognize the characters first, then the extraction step can rebuild useful document structure.
- Headings: Visible titles become # headings
- Paragraphs: Line fragments join into editable text
- Lists: Bullets and checklists keep item boundaries
- Tables: Simple grids become Markdown pipe tables
- Numbers: Dates and amounts remain reviewable text
- Source: The image stays beside the Text viewer
Extract first, then decide what to keep
- Upload and extract one document image without creating an account.
- Review low-confidence values beside the source document.
- Copy and Markdown download are available from the Text viewer after conversion.
- Hobby and above add stored files, batch jobs and connected exports. Email intake requires Pro or Ultra. File and processing limits depend on your plan.
Compare batch-processing plans or read how extraction and review work.
Check every character that carries meaning
OCR mistakes are often plausible. Compare names, dates, decimal points, identifiers, punctuation, and narrow table columns with the source. Retake a photo when blur, glare, perspective, or cropping hides part of the page.
Cases that need manual review
- Blur, glare, low contrast, aggressive JPEG compression, and very small text can replace or drop characters during OCR.
- A screenshot cropped through a heading, label, or table edge removes context that no parser can reliably reconstruct.
- Handwriting, formulas, diagrams, overlapping annotations, and dense multi-column pages may need substantial manual correction.
- A plausible character substitution in a name, date, amount or code may read naturally and still change the meaning.
- Multi-column screenshots can interleave paragraphs when the reading order is unclear, even if individual words are recognized.
- Turn a document screenshot into editable notes
- Recover text from a photographed page
- Convert a scan for use in a Markdown knowledge base
- Check a small table before copying it into documentation
Read the image OCR guide for photo, screenshot, scan, and table quality before choosing an output format.
Choose the page for the job
Use Image to Markdown for a JPG, PNG, screenshot, photo or scan whose visible writing should become editable prose. Capture the full page with good focus, even lighting and minimal perspective. After extraction, read the Markdown from beginning to end and check heading hierarchy, paragraph order, list nesting, names, punctuation, identifiers and simple tables against the image. Use PDF to Markdown when a PDF already has selectable text and can stay entirely inside the browser. Choose Image to Excel or Image to CSV when the main result is a table whose cells and numeric types matter. Choose Image to JSON when software needs named fields and row arrays. Markdown is a text structure, not a pixel-perfect reproduction, so formulas, diagrams, merged table cells and spatial annotations may need another workflow or manual editing. Image OCR may send content to the selected provider; check provider and policy settings before using confidential material.
Questions about Free Image to Markdown
The 28-day Free trial includes 25 page-equivalents per month shared by hosted and own-key processing, 5 hosted pages per document, and 5 documents per day. Anonymous trials have separate daily limits.
Is Image to Markdown browser-only?
No. Images need OCR because they have no embedded text layer. This workflow may send the image to the provider selected in the workspace. The page states that difference before upload.
Which image formats can I use?
The tool accepts JPG, JPEG, PNG, WebP, GIF, TIFF, and AVIF. The original image usually works better than a compressed copy from email or a messaging app.
Can it preserve a table?
It can express a clear, simple grid as a Markdown pipe table. Merged cells, nested headers, faint rules, and severe perspective distortion can need manual correction.
When should I use PDF to Markdown instead?
Use PDF to Markdown when the PDF already has selectable text. That path runs locally with no AI call. Use Image to Markdown for scans, photos, and screenshots that require OCR.
What should I review before reusing the Markdown?
Read the complete output in order. Compare names, dates, amounts, codes and punctuation with the image, then check heading levels, list nesting, paragraph sequence and table cells. If blur or cropping hides source text, retake the image rather than guessing the missing content.