MinerU document ingestion

MinerU is an open-source document conversion toolkit for turning PDFs and Office files into LLM-ready Markdown or JSON. Why it matters: Reliable document conversion is a bottleneck for retrieval, research agents, and knowledge-base ingestion. A maintained converter can reduce manual cleanup before content reaches an LLM.

Try this: Run it on a disposable batch of mixed PDFs, compare Markdown structure and citation preservation against the current conversion method, and inspect failures before automating ingestion.