MinerU vs Donut
Pick MinerU when you need full-page conversion of PDFs, images and Office files into Markdown, HTML, LaTeX or JSON without training anything. Pick Donut when you have a narrow, fixed-layout document type such as receipts or forms and can fine-tune it for field extraction. The main split is that MinerU works out of the box on whole documents, while Donut's base checkpoint is pretrain-only and must be fine-tuned first.
MinerUopendatalab/mineru | Donutnaver/donut | |
|---|---|---|
| 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 | 1.0 s / page |
| Best for | Academic PDFs | Receipt and form fields after fine-tuning |
| Inputs | PDF, images | images |
| Languages | 80+ languages | English, Chinese, Japanese, Korean |
| Max pages per request | 500 | 100 |
| Tables | ✓ Yes | No |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | No |
| Confidence | No | No |
| Forms | No | ✓ Yes |
| Multilingual | ✓ Yes | No |
| 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 accepts PDF, images, DOCX, PPTX, XLSX, EPUB, OFD and HTML in one tool, so mixed corpora for RAG need a single pipeline.
- It renders to Markdown, HTML, LaTeX, DOCX, EPUB and JSON, which suits full-page conversion of Chinese and English academic papers.
- Its basic tier runs on CPU with about 2 GB RAM, so you can start batch conversion without a GPU.
Choose Donut if
- It is MIT licensed and available in Hugging Face Transformers, which avoids the custom licence and AGPL-tagged weights that complicate MinerU in commercial use.
- It maps document images straight to structured JSON with no external OCR step, which fits receipt and form field extraction after fine-tuning.
- A practitioner on Hacker News reports that about 200 fine-tuning documents can reach human-level accuracy on narrow data.
Key differences
| MinerU | Donut | |
|---|---|---|
| Licence | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights are tagged AGPL-3.0 | MIT |
| Ready to use | Works as a toolkit with CLI, API or WebUI | Base checkpoint is pretrain-only and must be fine-tuned before use |
| Output formats | Markdown, HTML, LaTeX, DOCX, EPUB, JSON | JSON, plain text |
| Input types | PDF, images, DOCX, PPTX, XLSX, EPUB, OFD, HTML | Document images |
| Hardware | Basic tier: CPU with 2 GB RAM. VLM tiers: 16 GB RAM and a GPU with 8 GB+ VRAM | Not published |
| Benchmarks | olmOCR-Bench overall 75.2 (MinerU 2.5.4); ParseBench mean 45.9 (MinerU2.5) | Not published |
| Known failure | VLM tiers drop merged header rows in tables such as bank statement account lines; VLM pipeline can produce repeated content | Misses tiny text in large images due to input resolution limits; Transformers v5 dropped image-to-text pipeline support, so it needs direct loading or v4 |
Sources
- MinerU GitHub README
- olmOCR-Bench table
- Issue #5554 table header rows
- donut-base model card
- Donut ECCV paper
- Hacker News thread
What each one is
MinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU detailsDonut
OCR-free document understanding model. Fine-tuned for receipts, forms and document Q&A.
Donut 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": ["naver/donut"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
MinerU alternativesDonut alternativesBest OCR models in 2026