OpenViSTAI · Agents · Blockchain · Crypto
Submit a project
Menu
AI Models

Texo

Texo is a compact OCR model that converts images of mathematical expressions into LaTeX. Its public Python repository includes an inference notebook and a training pipeline, while the project also links to a browser-based demo.

Explore the practical guide ↓

Independent OpenViST project profile, based on linked sources.

Texo — OpenViST project profile in AI Models.
OpenViST project illustration. See the linked sources for product details.
Understand the project

Practical overview

When it fits

Use the online demo for a quick look at image-to-LaTeX recognition; use the repository when you need to inspect inference code, reproduce training, or study the model pipeline.

Who it is for

  • Students and researchers who need to transcribe mathematical notation from images.
  • Developers exploring formula OCR for note-taking software or a self-hosted service.
  • Machine-learning learners studying a LaTeX OCR model and its training process.

Documented uses

  • Recognize mathematical expressions in an image and produce LaTeX.
  • Explore inference through the online demo or repository notebook.
  • Train or resume training the model using the documented pipeline.
Before you choose

Requirements & limits

The repository describes Texo as free and open source and lists the AGPL-3.0 license. It links to an online demo.

  • The project provides both an online demo and a Python repository, supporting a quick trial as well as technical exploration.
  • The README lists 20 GB CPU memory and 16 GB GPU memory as minimum training resources when using streaming data loading and gradient accumulation; its recommendation is 50 GB CPU memory and 40 GB GPU memory.
  • The repository reports evaluation on UniMERNet-Test and cautions that results using its customized tokenizer are not directly comparable with the other listed metrics.
  • The project is licensed under AGPL-3.0.

Check current requirements in the official documentation. Source-based guidance does not mean OpenViST has run or independently tested this project.

Source-based practical guide

Set up Texo and download model weights

Prepare the Texo repository environment and fetch the model weights for later inference.

Before you start

  • Git
  • uv, as used in the repository setup instructions

Steps

  1. Clone the repository with the command below.

    Documented command
    git clone https://github.com/alephpi/Texo

  2. Install the project environment with the command below.

    Documented command
    uv sync

  3. Download the model with the command below.

    Documented command
    python scripts/python/hf_hub.py pull

  4. For training checkpoints, use the command below instead.

    Documented command
    python scripts/python/hf_hub.py pull --with_useful_ckpts

Expected outcome

The repository environment is configured and model files have been requested for local use.

Guide limitations

  • The repository points to demo.ipynb for inference but does not describe the inference steps here.
  • The guide documents setup and model download only; it does not establish that inference was run successfully.
Documentation 1 ↗
Evidence and transparency

Sources & review

Official repository ↗
Source record details
Source availability
Active
Sources checked
14 October 2026
Original website title
alephpi/Texo · Official homepage title · retained in its source language
How we review sourcesSuggest a correction