document OCR / parsing
Baidu's Unlimited-OCR is a vision-language model for one-shot, long-horizon document parsing and OCR.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
Same class of VLM document parser; pick if you prefer a model with ACL-2025 anchor prompting and can check its license.
The official repo for “Dolphin: Document Image Parsing via Heterogeneous Anchor Prompting”, ACL, 2025.
VLM multilingual layout parser; pick for an alternative model with different language strengths.
Multilingual Document Layout Parsing in a Single Vision-Language Model
VLM OCR model from a different lab; choose for its own speed/accuracy, but verify output format.
GLM-OCR: Accurate × Fast × Comprehensive
Lightweight VLM document parser; choose for lower inference resource requirements.
A lightweight LMM-based Document Parsing Model
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
OCR-free encoder-decoder; choose if you want a smaller model that skips explicit OCR tokens.
Official Implementation of OCR-free Document Understanding Transformer (Donut) and Synthetic Document Generator (SynthDoG), ECCV 2022
Detection-recognition pipeline toolkit; choose if you need 100+ language support and deployable tools.
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
End-to-end document conversion library; choose if you need layout analysis and multiple output formats.
Get your documents ready for gen AI
Document ETL pipeline; choose if you want connectors and chunking for LLM ingestion.
Convert documents to structured data effortlessly. Unstructured is open-source ETL solution for transforming complex documents into clean, structured formats for language models. Visit our website to learn more about our enterprise grade Platform product for production grade workflows, partitioning, enrichments, chunking and embedding.
Per-page vision LLM extraction; choose if you want to use cloud vision APIs rather than host a model.
OCR & Document Extraction using vision models
Rust-native document parser; choose for low-latency parsing in a Rust environment.
A fast, helpful, and open-source document parser
Trainable deep-learning OCR library; choose if you want to customize detection/recognition components.
docTR (Document Text Recognition) - a seamless, high-performing & accessible library for OCR-related tasks powered by Deep Learning. Ongoing development and maintenance by t2k.
Multi-format parsing wrapper; choose if you need a unified ingestion tool for many media types.
Ingest, parse, and optimize any data format ➡️ from documents to multimedia ➡️ for enhanced compatibility with GenAI frameworks
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
Classic lightweight OCR engine; choose for fast, offline extraction without deep learning dependencies.
Tesseract Open Source OCR Engine (main repository)
Original model this seed extends; choose for a known baseline or the older release.
Visual Causal Flow
Narrower: only works on digital PDFs, no OCR for scanned documents.
PDF Parser for AI-ready data. Automate PDF accessibility. Open-source.
Equation-specific OCR; choose only for converting math images to LaTeX.
pix2tex: Using a ViT to convert images of equations into LaTeX code.
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
End-user Android app with OCR; choose for interactive mobile editing, not programmatic use.
🖼️ Image Toolbox is a powerful app for advanced image manipulation. It offers dozens of features, from basic tools like crop and draw to filters, OCR, and a wide range of image processing options
Desktop screenshot app with OCR; choose for GUI capture/OCR, not an embeddable model.
ShareX is a free and open-source application that enables users to capture or record any area of their screen with a single keystroke. It also supports uploading images, text, and various file types to a wide range of destinations.
These projects were analysed and named Unlimited-OCR among their alternatives. The relationship is not symmetric — how Unlimited-OCR rates them is a separate judgement, made when Unlimited-OCR is analysed in its own right.
calls Unlimited-OCR “choose when you need one-shot parsing of very long documents beyond the seed's input length.”
Multilingual Document Layout Parsing in a Single Vision-Language Model
calls Unlimited-OCR “end-to-end OCR model, choose for one-shot long-horizon document parsing”
Official code implementation of General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
calls Unlimited-OCR “LMM-based one-shot long-horizon OCR, alternative document parser”
A lightweight LMM-based Document Parsing Model
calls Unlimited-OCR “Use for one-shot long-horizon parsing of documents into text/data instead of screenshot retrieval.”
The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/
calls Unlimited-OCR “if you need a one-shot long-horizon OCR model for complex documents”
Get your documents ready for gen AI
calls Unlimited-OCR “choose it for Baidu's long-horizon parsing model for whole-document OCR, if you want research code over a production platform.”
Ingest, parse, and optimize any data format ➡️ from documents to multimedia ➡️ for enhanced compatibility with GenAI frameworks