Donut vs OpenOCR
Pick OpenOCR if you need a working self-hosted parser for Chinese and English academic PDFs with formulas, since OpenDoc-0.1B reports 90.57 on OmniDocBench v1.5 at 0.1B parameters. Pick Donut if you have a narrow, fixed-layout document type such as receipts or forms and can fine-tune a model on your own samples. Donut ships as a pretrain-only base checkpoint, so it does nothing useful until you train it.
Donutnaver/donut | OpenOCRfudan/openocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 1 credit / page | 1 credit / page |
| Per 1,000 pages | $1.00 | $1.00 |
| Typical speed | 1.0 s / page | 0.5 s / page |
| Best for | Receipt and form fields after fine-tuning | Lightweight OCR pipeline on CPU |
| Inputs | images | PDF, images |
| Languages | English, Chinese, Japanese, Korean | Chinese and English |
| Max pages per request | 100 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | No |
| Forms | ✓ Yes | No |
| Multilingual | No | 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 Donut if
- You extract fields from receipts or forms with a fixed layout, and you can fine-tune on your own labelled samples.
- You want document images mapped straight to JSON with no separate OCR step.
- You want an MIT-licensed model that loads through Hugging Face Transformers, and you can pin Transformers v4 or load the model directly.
Choose OpenOCR if
- You parse Chinese and English academic PDFs that contain formulas, and you want LaTeX output.
- You want a small model: OpenDoc-0.1B and UniRec-0.1B are both 0.1B parameters.
- You need ONNX export for deployment and want bounding boxes in the output.
Key differences
| Donut | OpenOCR | |
|---|---|---|
| Licence | MIT | Apache-2.0 |
| How you run it | Self-hosted base model from Hugging Face. You fine-tune it for parsing, classification or VQA. | Self-hosted Python library covering detection, recognition, formula, table and document parsing. |
| Out-of-the-box use | Not usable without fine-tuning. The base checkpoint is pretrain-only. | Ready-made models: OpenDoc-0.1B for document parsing and UniRec-0.1B for text and formulas. |
| Output formats | JSON, plain text | Plain text, LaTeX, bounding boxes |
| Benchmark | Not published | 90.57 on OmniDocBench v1.5 (OpenDoc-0.1B), reported by the authors and not independently replicated |
| Languages | Not published | OpenDoc and UniRec support only Chinese and English |
| Known failure | Misses tiny text in large images because of input resolution limits. One HN user found results on personal documents did not match the promised scores. | Document parsing depends on a separate layout model, PP-DocLayoutV2 from PaddlePaddle. The README gives no hardware requirements or throughput figures. |
Sources
What each one is
Donut
OCR-free document understanding model. Fine-tuned for receipts, forms and document Q&A.
Donut detailsOpenOCR
Fudan's open toolkit for text detection and recognition, formulas, tables and document parsing.
OpenOCR 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": "naver/donut","fallbacks": ["fudan/openocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Donut alternativesOpenOCR alternativesBest OCR models in 2026