# MinerU | OpenDataLab | Knowledge Base | Instagit

Transforms complex documents like PDFs into LLM-ready markdown/JSON for your Agentic workflows.

GitHub Stars: 54.8k

Repository: https://github.com/opendatalab/mineru

---

## Articles

### [How to Deploy MinerU for Production Integration: Complete Setup Guide](/opendatalab/mineru/mineru-deployment-options-integration)

Deploy MinerU for production integration. Set up using Docker, FastAPI, OpenAI-compatible server, or Gradio to expose REST endpoints for document processing and retrieval.

- Tags: how-to-guide
- Published: 2026-02-23

### [MinerU Backend Source Files: Complete Guide to VLM, Pipeline, and Hybrid Locations](/opendatalab/mineru/mineru-backend-source-file-locations)

Discover the location of MinerU backend source files within the mineru/backend/ directory. Explore VLM, Pipeline, and Hybrid sub-packages including utils.py.

- Tags: how-to-guide
- Published: 2026-02-23

### [How to Set the Language for OCR in MinerU: CLI, Python API, and HTTP Methods](/opendatalab/mineru/mineru-configure-ocr-language)

Easily set the OCR language in MinerU using CLI, Python API, or HTTP requests. Learn how to specify languages for accurate text recognition with Pytorch-Paddle-OCR.

- Tags: how-to-guide
- Published: 2026-02-23

### [How to Specify the Parsing Method (Auto, TXT, OCR) in MinerU](/opendatalab/mineru/mineru-configure-parsing-method)

Control MinerU parsing with auto txt or ocr methods via CLI API or FastAPI. Learn how to specify your desired parsing method for efficient document analysis.

- Tags: how-to-guide
- Published: 2026-02-23

### [MinerU OmniDocBench Accuracy: Benchmark Results for Pipeline and VLM Backends](/opendatalab/mineru/mineru-accuracy-comparison-omnidocbench)

Discover MinerU OmniDocBench accuracy. See how MinerU's pipeline and VLM backends outperform GPT-4o and Gemini 2.5 Pro with just 1.2B parameters. Get benchmark results now.

- Tags: performance
- Published: 2026-02-23

### [How MinerU Detects and Handles Hallucinations in PDF Parsing](/opendatalab/mineru/mineru-hallucination-detection-mitigation)

MinerU prevents PDF parsing hallucinations by classifying documents. Learn how it avoids false text generation for corrupted or image-heavy files. Read more.

- Tags: how-to-guide
- Published: 2026-02-23

### [How to Use MinerU in HTTP Client Mode with OpenAI-Compatible Servers](/opendatalab/mineru/mineru-http-client-mode-openai-servers)

Learn to use MinerU in HTTP client mode with OpenAI-compatible servers. Offload VLM inference to remote servers for efficient processing.

- Tags: how-to-guide
- Published: 2026-02-23

### [VRAM Requirements for MinerU's VLM Backend: Complete Hardware Guide](/opendatalab/mineru/mineru-vlm-backend-vram-requirements)

Discover MinerU VLM backend VRAM requirements. Learn the minimum 8 GB VRAM needed and the recommended 10 GB for full table acceleration. Get your hardware guide now.

- Tags: hardware-guide
- Published: 2026-02-23

### [How to Utilize MinerU's VLM Backend for High-Accuracy Parsing](/opendatalab/mineru/mineru-vlm-backend-usage-accuracy)

Master PDF parsing with MinerU's VLM backend. Configure environment variables, choose an inference engine, and use MagicModel for accurate formula and table extraction, delivering structured Content-List V2 output.

- Tags: how-to-guide
- Published: 2026-02-23

### [How to Convert PDF to JSON Using MinerU: A Complete Technical Guide](/opendatalab/mineru/mineru-pdf-to-json-conversion)

Learn to convert PDF to JSON with MinerU. This technical guide details the six-stage pipeline for structured data extraction from PDFs using MagicModel inference. Explore the opendatalab MinerU repository.

- Tags: how-to-guide
- Published: 2026-02-23

### [How to Convert PDF to Markdown Using MinerU: A Complete Technical Guide](/opendatalab/mineru/mineru-pdf-to-markdown-conversion)

Easily convert PDF to Markdown with MinerU. This technical guide details MinerU's multi-stage pipeline for accurate document conversion to clean Markdown.

- Tags: how-to-guide
- Published: 2026-02-23

### [Complete Guide to Output Formats Supported by MinerU](/opendatalab/mineru/mineru-supported-output-formats)

Explore MinerU's seven output formats: Multimodal Markdown, NLP Markdown, JSON, raw output, images, and ZIP. Discover the best format for your data analysis.

- Tags: api-reference
- Published: 2026-02-23

### [How MinerU Handles Mathematical Formulas in PDFs: Detection, Recognition, and LaTeX Conversion](/opendatalab/mineru/mineru-formula-recognition-pdf)

Learn how MinerU detects, recognizes, and converts mathematical formulas in PDFs to LaTeX. Explore its advanced ViT-based recognition and efficient Markdown integration.

- Tags: how-to-guide
- Published: 2026-02-23

### [How MinerU Extracts Tables from PDFs: A Deep Dive into the Pipeline](/opendatalab/mineru/mineru-table-recognition-pdf)

Discover how MinerU extracts tables from PDFs. Explore its three-stage pipeline: rasterization, wired/wireless table classification with PaddlePaddle, and extraction via UNet or RapidTable models.

- Tags: deep-dive
- Published: 2026-02-23

### [How to Enable or Disable OCR in MinerU: Complete Configuration Guide](/opendatalab/mineru/mineru-configure-ocr)

Easily enable or disable OCR in MinerU via CLI, HTTP API, or Gradio UI. Configure OCR settings quickly with our comprehensive guide.

- Tags: how-to-guide
- Published: 2026-02-23

### [How MinerU Performs OCR on Scanned PDFs: A Deep Dive into the Pipeline](/opendatalab/mineru/mineru-ocr-functionality-scanned-pdfs)

Discover how MinerU performs OCR on scanned PDFs. Learn about its pipeline for text extraction, region detection, and layout integration. Explore the opendatalab/MinerU repository for details.

- Tags: deep-dive
- Published: 2026-02-23

### [System Requirements for Running MinerU: Hardware, Software, and GPU Configuration](/opendatalab/mineru/mineru-system-requirements)

Discover MinerU system requirements. Learn about essential hardware, software, and GPU configuration including RAM and VRAM needs for optimal performance. Run MinerU efficiently.

- Tags: system-requirements
- Published: 2026-02-23

### [How to Build and Install MinerU From Source: A Complete Developer Guide](/opendatalab/mineru/mineru-installation-source)

Build and install MinerU from source with our developer guide. Clone the opendatalab/MinerU repo, set up a Python 3.10–3.13 virtual environment, and install MinerU using uv pip.

- Tags: how-to-guide
- Published: 2026-02-22

### [How to Install MinerU Using pip: Complete Installation Guide](/opendatalab/mineru/mineru-installation-pip)

Easily install MinerU using pip. Get the full package with all optional backends by running pip install mineru[all] or install the lightweight core only with pip install mineru. Follow our complete installation guide.

- Tags: getting-started
- Published: 2026-02-22

### [How to Install MinerU Using Docker: Complete Setup Guide](/opendatalab/mineru/mineru-installation-docker)

Install MinerU with Docker easily. Build the GPU-optimized image and deploy OpenAI-compatible server, REST API, or Gradio UI. Follow our complete setup guide.

- Tags: how-to-guide
- Published: 2026-02-22

### [How to Use MinerU's Python API for Programmatic PDF Parsing](/opendatalab/mineru/mineru-python-api-usage)

Easily parse PDFs programmatically with MinerUs Python API. Utilize do_parse and aio_do_parse functions for seamless document extraction without the command line.

- Tags: how-to-guide
- Published: 2026-02-22

### [How MinerU's Hybrid Backend Combines Pipeline and VLM Advantages for PDF Parsing](/opendatalab/mineru/mineru-hybrid-backend-explained)

Discover how MinerU's hybrid backend leverages VLM and OCR pipelines to effectively parse PDFs, merging VLM layout understanding with OCR accuracy for unified JSON output.

- Tags: architecture
- Published: 2026-02-22

### [How to Configure MinerU for GPU, NPU, or CPU Environments](/opendatalab/mineru/mineru-hardware-configuration-gpu-npu-cpu)

Learn to configure MinerU for GPU, NPU, or CPU on macOS, Windows, or Linux. Control device selection with CLI flags or environment variables for optimal performance.

- Tags: how-to-guide
- Published: 2026-02-22

### [How to Choose Between MinerU's Pipeline, VLM, and Hybrid Backends: A Complete Guide](/opendatalab/mineru/mineru-backend-comparison-pipeline-vlm-hybrid)

Choose the right MinerU backend for your needs. Understand when to use pipeline, VLM, or hybrid for CPU, low-VRAM, or high-end GPU environments with VLM accuracy.

- Tags: how-to-guide
- Published: 2026-02-22