Docling
Docling is an easy-to-use, self-contained, MIT-licensed open source toolkit for converting messy documents into structured data and simplifying downstream document and AI processing. It can parse many popular document formats into a unified and richly structured Docling Document, including PDF, DOCX, PPTX, XLSX, HTML, Markdown, AsciiDoc, CSV, images, audio, and scanned pages through an OCR engine of the user’s choice. Docling detects tables, formulas, reading order, chunks, bounding boxes, page headers and footers, pictures, captions, code, list items, paragraphs, cells, and document structure, making extracted content easier to process, search, and ingest into AI, RAG, and agentic systems. It can export parsed documents to JSON, text, Markdown, HTML, and Doctags, giving developers flexible outputs for pipelines and applications. Docling stores and traverses components according to reading order, partitions documents into bite-sized contiguous text chunks.
Learn more
Upstage Document Parse
Upstage Document Parse transforms complex documents, PDFs, scanned images, spreadsheets, and slides containing text, tables, charts, and even handwriting, into structured, machine‑readable HTML or Markdown with enterprise‑grade speed and accuracy. Leveraging advanced layout understanding, it recognizes complex tables, charts, and element coordinates, processes pages at an average of 0.6 seconds each (100 pages in under a minute, 5–10× faster than competitors), and delivers over 5% higher layout and table recognition accuracy (TEDS: 93.48, TEDS‑S: 94.16). Easily invoked via a REST API or deployed on‑premises or through marketplaces like AWS, it fits seamlessly into existing pipelines using simple client libraries. Use cases span retrieval‑augmented enterprise search, AI‑powered document summarization, legal and compliance digitization, and financial report processing, preserving intricate layouts and ensuring clean, searchable outputs for downstream LLM workflows.
Learn more
IRISmart File
Semi-automatic naming and classification of your files: sort your stacks of documents efficiently into specific predefined folders in no time! (local or Cloud storage). Automatic creation, on the fly, of file storage tree structures, based on the root name of documents for easy and efficient filing. Process your documents at speeds of up to 30 pages per minute; includes the ability to rename, file, index, and compress a significant amount per day by running parallel batches in the background. Automatic separation of your various documents with blank pages: during scanning, insert blank pages between the documents you want to separate. IRISmart detects and creates automatic breaks exactly where you put them! Optimised archiving and searching via automatic conversion of your compressed and indexed PDF documents on the fly, while maintaining perfect legibility of the text.
Learn more
Restructured
Restructured is an AI-powered platform designed to help businesses extract insights from unstructured data at scale. Whether dealing with documents, images, audio, or video, it combines LLM capabilities with advanced search and retrieval methods to not only index information but also understand it in context. Restructured transforms massive datasets into actionable insights, making complex data easy to navigate and analyze.
Learn more
