MinerU

MinerU is a document reader that parses files and images into structured content for agent workflows.

MinerU is a document reading and parsing tool designed for AI agents and local workflows. It converts PDFs, scanned images, Office files, EPUBs, and web archives into structured formats like Markdown, HTML, and LaTeX. The tool supports multiple parsing tiers ranging from fast flash previews to advanced model-based layouts for tables and formulas. It features a document library, content search, and stable page or block locators for precise citations. Users can run it locally via a command-line interface, Python SDK, API, or Gradio web interface.

MinerU at a glance

Best for: Software engineer, Researcher
Runs on: CLI, Web, API, self-hosted, open source
Pricing: Free
Released: February 2024
Category: AI productivity tool
Website: github.com/opendatalab/MinerU

Key features

Parse PDFs, images, and Office documents
Extract tables, formulas, and complex layouts
Support multiple parsing tiers from flash to advanced
Provide stable page and block citation locators
Search and cache indexed local documents
Export to Markdown, HTML, LaTeX, and JSON

Use it for

Read and summarize local PDF and Office documents
OCR scanned documents and images for analysis
Convert documents into Markdown for LLM workflows
Extract structured tables and formulas from reports

Alternatives to MinerU

Docling — Provides advanced PDF understanding and local execution for sensitive data.
docTR — Extracts text and layout details from documents using deep learning.
MarkItDown — Converts documents into LLM-friendly Markdown for better token efficiency.
Sparrow — Extracts structured data locally from documents without cloud dependencies.
DocETL — Extracts structured data and links knowledge using large language models.
deepdoctection — Orchestrates layout analysis, OCR, and model fine-tuning in PyTorch pipelines.
Knowhere — It preserves page-native structures and source links for accurate agent retrieval.
LLM Wiki Compiler — Compiles raw sources into a citation-traced, interlinked markdown wiki.
LLM Wiki Agent — Compiles raw sources into a structured wiki that compounds over time.
NuMind — Achieves high accuracy in document extraction and outperforms generalist models.
OpenDataLoader PDF — It ranks first in benchmarks for layout and table extraction without needing a GPU.
IWE — Works natively across Mac apps, text editors, terminal CLI, and MCP servers.

All MinerU alternatives