> ## Documentation Index
> Fetch the complete documentation index at: https://parserport.outerport.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Open Source Tools

> Some useful pen source document processing tools and models.

## PDF Parsing Libraries

**[pypdfium2](https://github.com/pypdfium2-team/pypdfium2)**

* BSD-3 license
* Text extraction via `page.get_textpage()`
* Incremental parsing support
* Form XObjects support
* Production-ready performance

**[PyMuPDF (fitz)](https://pymupdf.readthedocs.io/)**

* GPL-3 license (commercial license available)
* Full text, image, vector graphics access
* Coordinate-mapped text extraction: `page.get_text("dict")`
* Vector graphics manipulation

**[pdfplumber](https://github.com/jsvine/pdfplumber)**

* MIT license
* Table extraction: `page.extract_table()`
* Configurable vertical/horizontal strategies
* SVG page preview generation

## Layout Analysis Models

**[DocLayout-YOLO](https://github.com/opendatalab/DocLayout-YOLO)**

* 18ms/page processing speed
* DocSynth-300K pre-training
* finetunable

**[PaddleOCR Layout](https://paddlepaddle.github.io/PaddleOCR/main/en/version3.x/module_usage/layout_detection.html)**

* PP-YOLOv2 / YOLOX architectures
* OCR pipeline integration

**[Deformable DETR](https://github.com/fundamentalvision/Deformable-DETR)**

* Multi-column layout support
* 2+ A100 GPU requirement
* Hugging Face/Detectron training

**[Grounding DINO + SAM](https://github.com/IDEA-Research/GroundingDINO)**

* Zero-shot text prompt detection
* 10-15 mAP improvement with fine-tuning

## Table Extraction Models

**[PaddleOCR Table](https://paddlepaddle.github.io/PaddleOCR/main/en/version3.x/module_usage/table_structure_recognition.html)**

* HTML output format
* COCO-style annotation support
* finetuable

**[Table Transformer (TATR)](https://github.com/microsoft/table-transformer)**

* Deformable DETR backbone
* PubTables-1M format support
* Row/column/cell detection

**[UniTable](https://github.com/poloclub/unitable)**

* Pixel-to-token framework that jointly predicts table structure, cell content & bounding boxes
* SOTA results on four benchmark datasets  
* MIT-licensed

**[Donut](https://github.com/clovaai/donut)**

* Direct image-to-markdown conversion
* 10k steps fine-tuning process
* Hugging Face integration

## General Purpose Libraries

**[PaddleOCR / PP-Structure](https://github.com/PaddlePaddle/PaddleOCR#pp-structure)**

* Layout, table, OCR, key-value pipeline
* YAML configuration system
* Standard training interface

**[Unstructured](https://docs.unstructured.io/welcome)**

* Element-based document partitioning
* Custom partitioner support

**[MarkItDown](https://github.com/microsoft/markitdown)**

* Multi-format (PDF, Office, HTML, images, audio) → Markdown conversion
* One-line CLI (`markitdown my.pdf > out.md`) + Python API
* MIT-licensed, plug-in architecture

**[docTR](https://github.com/mindee/doctr)**

* Apache-2.0 license
* DBNet, CRNN, ViTSTR models
* Hugging Face checkpoint compatibility