Report Bug / Feature Request
PDF Tools PDF Summarizer Chat with PDF PDF OCR Free AI OCR AI PDF Editor PDF to Word PDF to Excel More →

AI OCR to Markdown

Commercial use OK 380+ models No watermark No sign-up needed
Model:
+ GPT-5, Claude, Gemini
Drop an image, screenshot, or scan and get back clean Markdown — headings stay headings, tables stay tables, lists stay lists. Powered by our self-hosted Unlimited-OCR model, tuned for layout-aware document parsing rather than a flat text dump. Best for structured pages: papers, invoices, forms, reports.

Drop an image to convert to Markdown

PNG, JPG, WebP up to 10MB — one page per image

Preview
~800 tokens
Markdown output 

        

    

    

    

        
    







    
What it's good for

    

        

            
            
Tables & forms

            
Invoices, spec sheets, and reports come back as real Markdown tables instead of a jumbled wall of numbers — paste straight into a doc or a RAG pipeline.

        

        

            
            
Papers & articles

            
Headings, sections, and lists keep their structure, so a scanned report becomes editable Markdown you can drop into a docs site or feed to an LLM.

        

        

            
            
Dense & multilingual

            
The model is built for dense, complex pages and handles Chinese, English, and more — useful where a plain OCR engine garbles tight layouts.

        

    


    

         Plain Text OCR
         About the model
         PDF to Markdown
         Get Tokens
    




                        
                        

                        
                        
                        
                        
                        

                            Advanced options
                            

                                

                                    
                                

                                

                                    
                                

                            

                        

                    

                    
                


                
                

                    

                        

                            

                                
Result

                                

                                    
                                    
                                    
                                

                            

                            

                        

                    

                


                
                

                    

                         Tokens running low. 
                        Get More Tokens
                    

                


                
                

                    

                         Want better results? Premium models (GPT-5, Claude, Gemini) deliver higher quality.
                        View Plans
                    

                


                
                

                    
                    

                        
❤️ Love Free.ai? Tell your friends!

                        

                            
                            
                            
                            
                            
                        

                        
                        
                        
Sign up to get a referral link and earn 30,000 tokens per friend.

                        
                    


                    

                    
                    
                    

                        

                            

                                Want more?
                                Sign up free for 30K tokens/day
                            

                            Sign Up Free
                        

                    

                    
                


                
                

                    
                    
Processing your request...

                    

                        

                    

                


                
                
Convert any image, screenshot, or scan into clean Markdown with free AI OCR. Preserves headings, tables, and lists — powered by the self-hosted Unlimited-OCR model.

            

        

    







    

        

            

                
How to Use AI OCR to Markdown

                

                    

                        

                            1
                        

                        
Enter your input

                        
Type text, upload a file, or describe what you want. No account needed.

                    

                    

                        

                            2
                        

                        
Click generate

                        
Our AI processes your request in seconds using the best open-source models.

                    

                    

                        

                            3
                        

                        
Download & share

                        
Download, copy, or share your result. Free for personal and commercial use.

                    

                

            

        

    








    

        

            

                
Use this tool via API

                
Automate this tool from your own code. OpenAI-compatible REST endpoint, Bearer-token auth, no extra SDK required. Token costs match the web interface.

                

                    API Documentation
                    Get API Key
                

            

            

                
curl -X POST https://api.free.ai/v1/pdf/extract/ \
  -H "Authorization: Bearer sk-free-..." \
  -H "Content-Type: application/json" \
  -d '{"file": "@document.pdf"}'

            

        

    










    

        
Related Free AI Tools

        

            
            

                
                    PDF Tools
                
            

            
            

                
                    PDF Summarizer
                
            

            
            

                
                    Chat with PDF
                
            

            
            

                
                    PDF OCR
                
            

            
            

                
                    Free AI OCR
                
            

            
            

                
                    AI PDF Editor
                
            

            
            

                
                    PDF to Word
                
            

            
            

                
                    PDF to Excel
                
            

            
        

    









    

        

            

                
AI OCR to Markdown — FAQ

                

                    
                    

                        

                            
                        

                        

                            
Upload an image, screenshot, or scan and get back clean Markdown — headings, tables, and lists keep their structure instead of collapsing into a flat text dump. It is powered by Unlimited-OCR, a self-hosted 3B model tuned for layout-aware document parsing, so the output is ready to paste into a docs site, a note app, or an LLM pipeline.

                        

                    

                    
                    

                        

                            
                        

                        

                            
Plain OCR (our /ocr/ tool, Tesseract) gives you the raw words in reading order — great for a quick copy-paste, but tables flatten and headings lose their hierarchy. OCR to Markdown understands the page layout and reconstructs it: a table becomes a Markdown pipe-table, a heading stays a heading, a bullet list stays a list. Use plain OCR for speed, this for structure.

                        

                    

                    
                    

                        

                            
                        

                        

                            
Unlimited-OCR (baidu/Unlimited-OCR), an open MIT-licensed 3B model built on DeepSeek-OCR. We run it on our own GPUs — nothing is sent to a third-party API. It is purpose-built for one-shot document parsing rather than single-line text recognition.

                        

                    

                    
                    

                        

                            
                        

                        

                            
Yes — tables come back as proper Markdown pipe-tables, which is the whole point of using this over plain OCR. Very complex nested or multi-column tables flatten more aggressively (a fundamental Markdown limitation), but standard invoices, spec sheets, and report tables convert cleanly.

                        

                    

                    
                    

                        

                            
                        

                        

                            
It is multilingual with particularly strong Chinese and English support, and handles many other scripts. Accuracy is best on clear, high-resolution pages; very low-resolution or heavily stylized text is harder, as with any OCR.

                        

                    

                    
                    

                        

                            
                        

                        

                            
That tool takes a whole PDF and runs IBM Granite-Docling page by page — best when you already have a multi-page PDF. AI OCR to Markdown takes a single image or screenshot and runs the larger Unlimited-OCR model, which is stronger on dense or complex single pages. Use the PDF tool for documents you have as PDFs; use this for screenshots, photos of pages, and individual scans.

                        

                    

                    
                    

                        

                            
                        

                        

                            
AI OCR to Markdown takes one image (PNG / JPG / WebP) at a time — one page per image. For multi-page PDFs, use /document/pdf-to-markdown/ (whole-PDF Markdown) or /ocr/document/ (searchable-PDF / text). If you only have a PDF page, export or screenshot the page and drop it here.

                        

                    

                    
                    

                        

                            
                        

                        

                            
If you do not want Markdown structure — just the raw text in reading order — pick Plain text. The model then runs in straight-OCR mode and returns the words without headings, tables, or list formatting. Markdown is the default because it is what makes AI OCR to Markdown distinct.

                        

                    

                    
                    

                        

                            
                        

                        

                            
It works on phone photos of paper, but clean, flat, well-lit shots convert best. Glare, heavy skew, and shadows hurt accuracy. If a photo comes out poorly, re-shoot it straight-on with even lighting, or scan at 200+ DPI.

                        

                    

                    
                    

                        

                            
                        

                        

                            
Images are processed on our GPUs and discarded — we do not train on your uploads. We log only a timestamp and token cost for billing, never the document content. Signed-in users get a generation-history entry (a short preview) which you can delete anytime from /account/?tab=history. Privacy policy at /privacy/.

                        

                    

                    
                    

                        

                            
                        

                        

                            
Free in the daily token pool (~800 tokens per page) — no credit card. Your first conversions work without an account; signing up unlocks the full daily pool, history, and access across all of Free.ai's 400+ tools.

                        

                    

                    
                    

                        

                            
                        

                        

                            
Yes. POST multipart to /v1/ocr/ with your image, model=unlimited-ocr, and output_format=markdown (or plain). The response JSON has a text field with the Markdown. Bearer auth, 10K tokens/month free. /api/ has the curl example.

                        

                    

                    
                

            

        

    









    

        
Sign up free — 30,000 tokens/day

        Create Free Account
        
No credit card required

    









    
How would you rate this tool?

    

        
        
        
        
        
        
        
        
        
        
        
    

    

        
    


    
    








    


    
    



    



    


    

        
Love Free.ai? Tell your friends!