Docling
Partial infoOpen-source document parsing library for RAG.
Docling is an open-source (MIT) library, originally from IBM Research and now a Linux Foundation AI project, that parses PDFs and other documents into structured formats (incl. tables and layout) for RAG and LLM pipelines. It runs locally.
License
Open source (MIT)
Deployment
Self-hosted
Pricing
Free and open-source; runs locally.
SDKs / languages
Python
Founded
2024
Strengths
- Free, open-source, local
- Good layout/table handling
- Strong backing (LF AI)
Limitations
- Library, not a managed service
- Python-only
Alternatives
Firecrawl
Turn websites into clean markdown for LLMs.
LlamaIndex
Data framework for RAG and agents over your data.
LlamaParse
High-fidelity parsing of complex PDFs for RAG.
Reducto
High-accuracy document parsing and extraction API.
Unstructured
Turn messy documents into clean, structured data for RAG.
Facts verified 19 July 2026. Neutral summary, not an endorsement; verify current details with the vendor.