Learn to Unlock Document Intelligence with Open-Source AI
summary
Docling is an MIT‑licensed Python library that actually converts 15+ file formats into a structured intermediate Docling document using OCR, layout models, Granite VLMs, Replicate/Hugging Face models and exports to markdown/HTML/JSON. The workshop shows chunking tuned to embedding windows, building a multimodal RAG with visual grounding (LangChain + Chroma + Granite embeddings) and extracting charts, tables and coordinates. Real numbers: a partner processes hundreds of millions of pages and reports processing 918M pages for $35,000 versus $750k for OCR only.
AI-written summary · about
chapters (8)
- 0:00Introduction
- 1:51What is Dockling and why it matters
- 10:25Document conversion basics
- 21:37Extracting tables and images
- 39:34Chunking for RAG
- 54:21Advanced features
- 1:00:21Multimodal RAG system
- 1:15:19Accuracy evaluation
related talks
01h 25mLearn to Unlock Document Intelligence with Open-Source AI
01h 00mComplex Data Ingestion with Open Source AI
01h 30mUnlock Information from Tables, Images and Complex Docs
01h 07mMeet Docling: The “Pandas” for Document AI
38mBuild AI-Ready Search: Integrating Docling with OpenSearch for Advanced RA...
48mIntegrating Docling With OpenSearch for Advanced RAG and Agentic Applications
31mDocling: Get Your Documents Ready for Gen AI
21mStructuring the Unstructured