Document Parsing & OCR

MinerU

Turn complex papers and reports into structured content, including text, tables, and formulas.

At a glance

Input
PDF and page images
Output
Markdown, JSON
Provenance
Page and region information
Scanned pages
Supported with the selected pipeline
Deployment
Self-hosted
Availability
Source available
Costs to plan for
Compute and model costs

Limitations to check

Check reading order in two-column pages, cross-page table boundaries, and the current code and model terms.

Build a meaningful sample

Include your hardest pages: low-quality scans, missing fields, dense tables, and evidence split across pages. Validate output correctness separately from citation accuracy.