MinerU vs RapidOCR
Pick MinerU when you need Markdown, HTML, LaTeX or DOCX out of PDFs, Office files and academic papers, and you can run a VLM on a GPU or accept basic quality on CPU. Pick RapidOCR when you only need text and bounding boxes from Chinese or English images and want a small offline library you can call from Python, C++, Java or C#. The main split is document parsing versus plain line OCR.
MinerUopendatalab/mineru | RapidOCRrapidai/rapidocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 2 credits / page | 1 credit / page |
| Per 1,000 pages | $2.00 | $1.00 |
| Typical speed | 2.0 s / page | 0.6 s / page |
| Best for | Academic PDFs | Lightweight OCR on CPU or edge |
| Inputs | PDF, images | images |
| Languages | 80+ languages | Chinese, English and more |
| Max pages per request | 500 | 100 |
| Tables | ✓ Yes | No |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | ✓ Yes |
| Forms | No | No |
| Multilingual | ✓ Yes | ✓ Yes |
| On OCRRouter | Available | Available |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
Which to choose
Choose MinerU if
- It takes PDF, images, DOCX, PPTX, XLSX, EPUB, OFD and HTML in one tool, which suits mixed corpora for RAG.
- It renders to Markdown, HTML, LaTeX, DOCX, EPUB and JSON, so you get structured output and not just text.
- It scored 75.2 on olmOCR-Bench (version 2.5.4), and the VLM tier reaches 2.12 pages per second on one A100 with vLLM.
Choose RapidOCR if
- It installs with pip install rapidocr onnxruntime and runs PaddleOCR models without the Paddle framework.
- It has bindings for Python, C++, Java and C#, so it embeds in desktop and server apps in many stacks.
- It runs on CPU through ONNX Runtime by default, with optional TensorRT, OpenVINO, MNN, Paddle or PyTorch engines.
Key differences
| MinerU | RapidOCR | |
|---|---|---|
| What it is | Document parsing toolkit that turns PDFs, images and Office files into Markdown or JSON | OCR library that runs converted PaddleOCR models for offline inference |
| Output | Markdown, HTML, LaTeX, DOCX, EPUB, JSON | Plain text and bounding boxes only |
| Hardware | Basic ONNX tier runs on CPU with 2 GB RAM; VLM tiers need 16 GB RAM and a GPU with 8 GB+ VRAM | CPU via ONNX Runtime by default; GPU and other engines are optional |
| Licence | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights tagged AGPL-3.0 | Apache-2.0; converted models keep their upstream licences |
| Benchmark | olmOCR-Bench 75.2 (MinerU 2.5.4); ParseBench mean 45.9 (MinerU2.5) | Not published |
| Languages | Good for Chinese and English academic papers | Chinese and English only by default; other languages need extra models |
| Known failure | VLM tiers drop merged header rows in tables such as bank statement account lines; the VLM pipeline can repeat content, producing 300 KB figure blocks | 16-bit grayscale PNG or TIFF give no text or wrong text; Arabic recognition crashes because python-bidi is undeclared |
Sources
- MinerU GitHub README
- olmOCR-Bench table
- Issue #5554 table header rows
- RapidOCR GitHub README
- Issue #776
- Issue #719
What each one is
MinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU detailsRapidOCR
PaddleOCR models packaged for ONNX Runtime and other engines. Easy to deploy without Paddle.
RapidOCR detailsTry both on your own file
Run the file once with each model and compare the output.
Use one, fall back to the other
{"model": "opendatalab/mineru","fallbacks": ["rapidai/rapidocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
MinerU alternativesRapidOCR alternativesBest OCR models in 2026