MinerU

概要

MinerU is an MCP server that integrates with the MinerU document parsing API, enabling robust extraction of structured content from diverse file types including PDFs, DOCs, and images. It supports both high-accuracy VLM models and fast pipeline models, offering features like batch processing for up to 200 documents, local file uploads, and specific page range extraction. Designed for developers and AI assistants, it streamlines the process of converting complex documents into usable markdown, optimized for efficient processing by models like Claude Code.

主な機能

• Batch processing for up to 200 documents via URLs or local file uploads
• Extract specific page ranges and support for 109 languages via OCR
• Optimized for Claude Code, reducing token usage by 73% vs alternatives
• Fast pipeline model for simple documents
• High-accuracy VLM model for complex documents
• 0 GitHub stars

ユースケース

• Automate document content extraction for data analysis or archiving
• Convert large batches of PDFs, DOCs, and images into structured markdown for AI processing
• Integrate advanced document parsing capabilities directly into AI coding environments and IDEs