@josesilesdata: This FREE OCR can read doctors' handwriting, complex tables, and mathematical equations. It's called **Chandra**, from …

X AI KOLs Timeline Tools
ocr handwriting-recognition table-extraction latex markdown open-source

Summary

Chandra is a free OCR tool from Datalab that can read doctors' handwriting, complex tables, mathematical equations, and forms, outputting results in Markdown, HTML, or JSON. It supports over 40 languages and can be installed via pip.

This FREE OCR can read doctors' handwriting, complex tables, and mathematical equations. It's called **Chandra**, from Datalab. And it's one of the most complete OCRs I've seen. It can process: **Handwritten text** Medical notes, filled-out forms, cursive or hard-to-read handwriting. **Complex tables** Preserves merged cells, rows, and columns. Ideal for invoices, financial reports, and PDFs with lots of data. **Mathematical formulas** Converts equations directly to LaTeX. **Forms** Recognizes checkboxes, buttons, and entered values. **Documents with tricky layouts** Multi-column text, newspapers, books, and documents with images. It can return the result in Markdown, HTML, or JSON. It works in over 40 languages. To install it: `pip install chandra-ocr` It's free for personal use. I'll leave the link to the repo in the comments
Original Article
View Cached Full Text

Cached at: 07/20/26, 09:30 AM

This FREE OCR can read doctors’ handwriting, complex tables, and mathematical equations.

It’s called Chandra, from Datalab.

And it’s one of the most complete OCRs I’ve seen.

It can process:

Handwritten text Medical notes, filled-out forms, cursive or hard-to-read handwriting.

Complex tables Preserves merged cells, rows, and columns. Ideal for invoices, financial reports, and PDFs with lots of data.

Mathematical formulas Converts equations directly to LaTeX.

Forms Recognizes checkboxes, buttons, and entered values.

Documents with tricky layouts Multi-column text, newspapers, books, and documents with images.

It can return the result in Markdown, HTML, or JSON.

It works in over 40 languages.

To install it:

pip install chandra-ocr

It’s free for personal use.

I’ll leave the link to the repo in the comments

Similar Articles

allenai/olmocr

GitHub Trending (daily)

olmOCR is an open-source toolkit from AI2 that converts PDFs, PNGs, and JPEGs into clean Markdown text, supporting equations, tables, and complex layouts. It includes a benchmark suite and multiple model releases (v0.4.0 as of Oct 2025) with improved performance and efficiency.