Docling

Docling is a document processing tool that parses diverse formats and integrates with generative AI.

Docling simplifies document processing by parsing diverse formats and providing integrations with the generative AI ecosystem. It offers advanced PDF understanding including page layout, reading order, table structure, code, formulas, and image classification. The tool supports multiple document formats such as PDF, DOCX, PPTX, XLSX, HTML, EPUB, audio, video, images, and email formats. It includes local execution capabilities for sensitive data and plug-and-play integrations with LangChain, LlamaIndex, Crew AI, and Haystack. Users can export data to Markdown, HTML, JSON, and other formats while using OCR and Visual Language Models.

Docling at a glance

Best for: Software engineer, Researcher
Runs on: macOS, Linux, Windows, CLI, API, self-hosted, open source
Pricing: Free
Released: July 2024
Category: AI productivity tool
Website: github.com/docling-project/docling

Key features

Parsing multiple document formats including PDF, DOCX, and images
Advanced PDF understanding for layout, tables, and formulas
Local execution capabilities for sensitive data
Plug-and-play integrations with LangChain and LlamaIndex
Extensive OCR support for scanned documents
Exporting to Markdown, HTML, and lossless JSON
Chart understanding to convert charts into tables or code

Use it for

Parse diverse document formats for generative AI applications
Extract table structures and layouts from complex PDFs
Process sensitive documents locally in air-gapped environments
Convert charts into tables or code with detailed descriptions

Alternatives to Docling

MinerU — Parses complex documents and tables into structured markdown for agents.
docTR — Extracts text and layout details from documents using deep learning.
PDF Craft — It preserves formulas, tables, and document structure during conversion.
DocETL — Extracts structured data and links knowledge using large language models.
MarkItDown — Converts documents into LLM-friendly Markdown for better token efficiency.
OpenDataLoader PDF — It ranks first in benchmarks for layout and table extraction without needing a GPU.
MarkItDown AI — Convert local documents directly in your browser without installing Python.
deepdoctection — Orchestrates layout analysis, OCR, and model fine-tuning in PyTorch pipelines.
MCP Docs — Runs locally, edits office files, and connects local AI coding agents directly.
PDF Inspector — Detects if a PDF is scanned or text-based to avoid unnecessary OCR costs.
MiniCPM OCR — Efficient document parsing with correctness assessment and high throughput.
NuMind — Achieves high accuracy in document extraction and outperforms generalist models.

All Docling alternatives