OmniParse

OmniParse: Universal Data Ingestion and Parsing Library for GenAI Frameworks

OmniParse is an open-source Python library designed to ingest, parse, and optimize arbitrary data formats—documents, multimedia, and more—for compatibility with generative AI frameworks. The project has accumulated 7,349 GitHub stars with 125 added today, indicating active community traction. It targets the data preprocessing layer of AI pipelines, a common friction point in RAG and agent workflows.