document OCR / parsing
PaddleOCR is a comprehensive OCR and document parsing toolkit that converts PDFs and images into structured, LLM-ready data (JSON/Markdown), supporting 100+ languages.
Same problem, same approach. Swapping one for another is a config change, not a rewrite.
when you prefer a parser focused exclusively on document-to-markdown rather than a general OCR toolkit.
Transforms complex documents like PDFs and Office docs into LLM-ready markdown/JSON for your Agentic workflows.
when you need a lightweight Python API that also emits CSV/HTML for downstream apps.
Extract and convert data from any document, images, pdfs, word doc, ppt or URL into multiple formats (Markdown, JSON, CSV, HTML) with intelligent structured data extraction and advanced OCR.
if you want a fast Rust-based parser with a small footprint and Apache-2.0.
A fast, helpful, and open-source document parser
if you need a Java/JVM parser integrated into JVM data pipelines.
PDF Parser for AI-ready data. Automate PDF accessibility. Open-source.
if you want an open-source toolkit with unified training/evaluation and reproducible OCR baselines.
OpenOCR: An Open-Source Toolkit for General-OCR Research and Applications, integrates a unified training and evaluation benchmark, commercial-grade OCR and Document Parsing systems, and faithful reproductions of the core implementations from a wide range of academic papers.
if you prefer a Node.js pipeline that outputs enriched JSON for downstream processing.
Transforms PDF, Documents and Images into Enriched Structured Data
when you want a full ETL platform with connectors, API, and pre-built chunking.
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.
when you want an AGPL-licensed modular PDF extraction toolkit with high-quality layout outputs.
A Comprehensive Toolkit for High-Quality PDF Content Extraction
Solves the same problem with a different architecture or at a different layer. Expect to rewrite the integration.
when you need deterministic rule-based parsing/chunking for RAG with an MCP server.
Fast, deterministic document parser and chunker for RAG and agent pipelines. Ships an MCP server for agents navigating long documents with precision.
when you want to use vision-language models for OCR and can call external model APIs.
OCR & Document Extraction using vision models
when you need a self-hosted document parsing server with HITL review and multi-format ingestion.
Privacy-first document intelligence engine — parse PDFs, DOCX, PPTX, XLSX & CSV into AI-ready chunks for RAG pipelines. Includes HITL review, 3-layer memory chat, and a production FastAPI server.
An earlier or more limited way to do the job. Still the right call when you need something small, proven, or CPU-only.
a research framework for diffusion-based OCR decoding, not a full document parser.
[ECCV 2026] A diffusion-based framework for document OCR that replaces autoregressive decoding with block-level parallel diffusion decoding.
when you only need table extraction and want a dedicated lightweight model.
A Unified Toolkit for Deep Learning-Based Table Extraction
when your task is key-info extraction from cards/bills using LLM prompts, not full-page OCR.
利用llm大语言模型提取卡证票据关键信息。Key Information Extraction from Image with LLM(large language model).Basically, it can extract key information from all bills and documents.
when you are transcribing historical handwritten/printed documents with specialized OCR models.
CHURRO is an OCR toolkit for historical document transcription, built to make handwritten and printed sources readable at high accuracy and lower cost.
Does the same job, but ships as an app. Useful to a person, not swappable into a codebase.
a Vue-based document parsing app/service from KDAN, not a code-embeddable library.
Part of the KDAN ecosystem, DocSlight offers document parsing, OCR, and data extraction that turn PDFs, scans, images, and Office files into structured outputs for RAG pipelines, AI agents, and enterprise document automation.
a web-based document parsing demo/app, not a library.
With DocSlight, precisely parse and extract data from any document, including PDFs, scans, images, and Office files. It is an open-source AI project from ComPDF (KDAN ecosystem).
a desktop OCR GUI app for end users, not a library for developers.
OCR software, free and offline. 开源、免费的离线OCR软件。支持截屏/批量导入图片，PDF文档识别，排除水印/页眉页脚，扫描/生成二维码。内置多国语言库。
These projects were analysed and named PaddleOCR among their alternatives. The relationship is not symmetric — how PaddleOCR rates them is a separate judgement, made when PaddleOCR is analysed in its own right.
calls PaddleOCR “active Python/PP-OCR toolkit with 100+ languages and PDF-to-structured-data; choose for broad language support plus OCR and document parsing.”
Tesseract Open Source OCR Engine (main repository)
calls PaddleOCR “Choose when you need a more production-ready OCR toolkit with built-in table/layout recognition and a larger model zoo, but you are comfortable with PaddlePaddle instead of PyTorch.”
Ready-to-use OCR with 80+ supported languages and all popular writing scripts including Latin, Chinese, Arabic, Devanagari, Cyrillic and etc.
calls PaddleOCR “drop-in peer with mature, widely-adopted OCR/layout/table toolkit and extensive language support.”
OCR, layout analysis, reading order, table recognition in 90+ languages
calls PaddleOCR “choose for up to 100+ languages and a mature Apache-2.0 OCR ecosystem.”
OCR model that handles complex tables, forms, handwriting with full layout.
calls PaddleOCR “Choose for a production-ready OCR toolkit with 100+ languages and extensive deployment options.”
GLM-OCR: Accurate × Fast × Comprehensive
calls PaddleOCR “heavyweight OCR toolkit with 100+ languages, good for image/PDF extraction but more resource-intensive.”
Transforms PDF, Documents and Images into Enriched Structured Data
calls PaddleOCR “Full-featured OCR toolkit with broad language support and pretrained models, a mature drop-in peer for OpenOCR.”
OpenOCR: An Open-Source Toolkit for General-OCR Research and Applications, integrates a unified training and evaluation benchmark, commercial-grade OCR and Document Parsing systems, and faithful reproductions of the core implementations from a wide range of academic papers.
calls PaddleOCR “When you need a battle-tested OCR engine with 100+ languages and full pipeline control.”
Convert PDF to markdown + JSON quickly with high accuracy
calls PaddleOCR “Choose if you need a battle-tested OCR library with hundreds of languages and fine control; it provides text boxes but not full Markdown/JSON conversion.”
PDF Parser for AI-ready data. Automate PDF accessibility. Open-source.
calls PaddleOCR “Detection-recognition pipeline toolkit; choose if you need 100+ language support and deployable tools.”
Unlimited OCR Works: Welcome the Era of One-shot Long-horizon Parsing.
calls PaddleOCR “OCR engine for images and scanned docs; choose when input is raster, not digital.”
Convert Word, PowerPoint, Excel, OpenDocument, RTF, EPUB, CSV, and PDF to clean Markdown. Built in Rust, with Node.js and Python bindings.
calls PaddleOCR “Choose if you need multi-language OCR and a traditional pipeline that can be customized for many document types.”
Toolkit for linearizing PDFs for LLM datasets/training
calls PaddleOCR “when you want an OCR-centric toolkit with huge language coverage and layout/table recognition rather than a full ETL pipeline.”
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.
calls PaddleOCR “when you need a powerful multi-language OCR toolkit with traditional layout analysis.”
OCR & Document Extraction using vision models
calls PaddleOCR “Use when you need mature OCR text extraction (100+ languages) from PDFs/images instead of a visual retrieval index.”
The end of web parsing. The beginning of scalable pixel-native search. link: https://pixelrag.ai/
calls PaddleOCR “choose when you need a modular, trainable OCR pipeline with CPU/edge deployment instead of a single VLM.”
Multilingual Document Layout Parsing in a Single Vision-Language Model
calls PaddleOCR “choose it for production-grade OCR in 100+ languages with strong on-prem performance, if you are willing to build the parsing pipeline yourself.”
Ingest, parse, and optimize any data format ➡️ from documents to multimedia ➡️ for enhanced compatibility with GenAI frameworks
calls PaddleOCR “Choose this original PaddleOCR framework when you need model training/fine-tuning and native PaddlePaddle support, at the cost of heavier dependencies.”
📄 Awesome OCR multiple programing languages toolkits based on ONNX Runtime, OpenVINO, MNN, PaddlePaddle, TensorRT and PyTorch.
calls PaddleOCR “classic OCR toolkit with broad language support, different architecture from LMM”
A lightweight LMM-based Document Parsing Model
calls PaddleOCR “Choose for an all-in-one OCR toolkit with more languages and layout awareness than Tesseract alone.”
Enhances Tesseract OCR output using LLMs (local or API) for error correction, smart chunking, and markdown formatting of scanned PDFs