# VikParuchuri/texify

Math OCR model that outputs LaTeX and markdown

Repository: https://github.com/VikParuchuri/texify
Canonical: https://ross.abutalabs.com/products/texify
Homepage: https://www.datalab.to
Language: Python
License: GPL-3.0
License Family: copyleft
Topics: latex, markdown, ocr, deep-learning
Archived: true
Last push: 2025-01-29T15:58:56+00:00

## Health v2 (maintenance only)
Score: 10/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 4, release rhythm 40, longevity 70
- inputs: {"age_days": 989, "days_push": 581, "days_rel": 671, "gap_med": 15, "n_releases_24m": 2}
- flags: archived
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1126, forks 92 (observed 2026-08-28T04:03:41.345365+00:00)

## What it is
Texify is a deep-learning OCR model that converts images or PDFs containing math into markdown and LaTeX renderable by MathJax. It handles both block equations and inline math mixed with text, running on CPU, GPU, or MPS.

## Use cases
- convert images of math equations to latex
- ocr math from pdfs into markdown
- extract latex from screenshots of equations
- digitize handwritten or printed math notation
- convert inline math and text from images
- build a mathpix snip alternative

## When to choose
- you need open-source math OCR that handles both equations and surrounding text
- you want LaTeX output compatible with MathJax delimiters
- you need to run OCR on CPU, GPU, or Apple MPS

## When to avoid
- starting a new project - the project is deprecated and migrated to surya
- you only need block equation OCR (pix2tex may suffice)
- you need full-page document OCR (nougat or surya are better fits)

## Facets
- artifact type: library
- maturity: abandoned
- function: ocr, machine-learning, deep-learning, pdf, image-processing
- domain: computer-vision, pdf, education
- platform: python, cli, cross-platform
- tags: latex-ocr, math-ocr, markdown-output, deprecated, mathjax, pytorch, natural-language-processing, gpu

## Member repositories
- VikParuchuri/texify (main) score 10

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:41.345365+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:39:11.123885+00:00, confidence not recorded.
  - readme: https://github.com/VikParuchuri/texify (fetched 2026-08-28T04:03:41.345365+00:00, sha edd647df26c5)
  - homepage: https://www.datalab.to (fetched 2026-08-29T12:43:42.690287+00:00, sha 2bbd3c9c83e2)
  - registry_pypi: https://pypi.org/pypi/texify/json (fetched 2026-08-29T12:43:42.699296+00:00, sha 1f1f7b78a118)
- Data as of 2026-08-30T08:39:29.467469+00:00.