Texify is an OCR model that converts images or pdfs containing math into markdown and LaTeX that can be rendered by MathJax ($$ and $ are delimiters). It can run on CPU, GPU, or MPS.
Texify can work with block equations, or equations mixed with text (inline). It will convert both the equations and the text.
Pix2tex is trained on im2latex, and nougat is trained on arxiv. Texify is trained on a more diverse set of web data, and works on a range of images.
See more details in the benchmarks section.
Discord is where we discuss future development.
Note I added spaces after _ symbols and removed , because Github math formatting is broken.
You'll need python 3.9+ and PyTorch. You may need to install the CPU version of torch first if you're not using a Mac or a GPU machine. See here for more details.
`pip install texify`
Model weights will automatically download the first time you run it.
texify/settings.py. You can override any settings with environment variables.
I've included a streamlit app that lets you interactively select and convert equations from images or PDF files. Run it with:
The app will allow you to select the specific equations you want to convert on each page, then render the results with KaTeX and enable easy copying.
You can OCR a single image or a folder of images with:
texify /path/to/folder_or_file --max 8 --json_path results.json
--max is how many images in the folder to convert at most. Omit this to convert all images in the folder.
--json_path is an optional path to a json file where the results will be saved. If you omit this, the results will be saved to
--katex_compatible will make the output more compatible with KaTeX.
You can import texify and run it in python code:
from texify.inference import batch_inference
from texify.model.model import load_model
from texify.model.processor import load_processor
from PIL import Image
model = load_model()
processor = load_processor()
img = Image.open("test.png") # Your image name here
results = batch_inference([img], model, processor)
texify/output.py:replace_katex_invalid if you want to make the output more compatible with KaTeX.
If you want to develop texify, you can install it manually:
git clone https://github.com/VikParuchuri/texify.git
poetry install # Installs main and dev dependencies
OCR is complicated, and texify is not perfect. Here are some known limitations:
Benchmarking OCR quality is hard - you ideally need a parallel corpus that models haven't been trained on. I sampled from arxiv and im2latex to create the benchmark set.
Each model is trained on one of the benchmark tasks:
Although this makes the benchmark results biased, it does seem like a good compromise, since nougat and pix2tex don't work as well out of domain. Note that neither pix2tex or nougat is really designed for this task (OCR inline equations and text), so this is not a perfect comparison.
|Edit Distance ⬇
You can benchmark the performance of texify on your machine.
pip install pix2tex
pip install nougat-ocr
benchmark.py like this:
python benchmark.py --max 100 --pix2tex --nougat --data_path data/bench_data.json --result_path data/bench_results.json
This will benchmark marker against pix2tex and nougat. It will do batch inference with texify and nougat, but not with pix2tex, since I couldn't find an option for batching.
--max is how many benchmark images to convert at most.
--data_path is the path to the benchmark data. If you omit this, it will use the default path.
--result_path is the path to the benchmark results. If you omit this, it will use the default path.
--pix2tex specifies whether to run pix2tex (Latex-OCR) or not.
--nougat specifies whether to run nougat or not.
Texify was trained on latex images and paired equations from across the web. It includes the im2latex dataset. Training happened on 4x A6000s for 2 days (~6 epochs).
This work would not have been possible without lots of amazing open source work. I particularly want to acknowledge Lukas Blecher, whose work on Nougat and pix2tex was key for this project. I learned a lot from his code, and used parts of it for texify.