Unlimited OCR vs Donut
Pick Unlimited OCR to turn long multi-page PDFs such as reports and books into Markdown with bounding boxes, with no training step. Pick Donut when you have a narrow, fixed-layout document type like receipts or forms and can fine-tune a model to output JSON fields. The main split is that Unlimited OCR works out of the box on whole documents, while Donut is a pretrain-only base that needs fine-tuning first.
Unlimited OCRbaidu/unlimited-ocr | Donutnaver/donut | |
|---|---|---|
| Type | Open source | Open source |
| Price | 2 credits / page | 1 credit / page |
| Per 1,000 pages | $2.00 | $1.00 |
| Typical speed | 1.2 s / page | 1.0 s / page |
| Best for | Books and very long documents | Receipt and form fields after fine-tuning |
| Inputs | PDF, images | images |
| Languages | 30+ languages | English, Chinese, Japanese, Korean |
| Max pages per request | 1,000 | 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 Unlimited OCR if
- It parses 40+ pages in a single forward pass with roughly constant memory, so long reports and books go through in one request.
- It outputs Markdown, plain text and bounding boxes, which suits layout-aware extraction and bulk PDF to text for RAG.
- It is usable without training, and Baidu Cloud offers a hosted endpoint if you do not want to run GPUs yourself.
Choose Donut if
- It maps a document image straight to structured JSON with no separate OCR step, which fits receipt and form field extraction.
- A practitioner on Hacker News reports that about 200 fine-tuning documents can reach human-level accuracy on narrow data.
- It suits document classification and fixed-layout document types once fine-tuned, and it loads in Hugging Face Transformers under the MIT licence.
Key differences
| Unlimited OCR | Donut | |
|---|---|---|
| Ready to use | Parses documents out of the box | Base checkpoint is pretrain-only and must be fine-tuned before use |
| Output formats | Markdown, plain text, bounding boxes | JSON, plain text |
| How you run it | Self-host with vLLM, SGLang or transformers; also hosted in Baidu Cloud | Self-host only |
| Licence and price | MIT, free self-hosted (you pay for compute) | MIT, free self-hosted (you pay for compute) |
| Hardware | NVIDIA GPU with CUDA 12.9 or 13.0; version 2 under SGLang runs out of memory on 12 GB GPUs | Not published |
| Benchmarks | ParseBench mean 46.17 (independent); Unlimited OCR 2 scores 84.6 on MPDocBench and 77.1 on MarathonDocBench (by the authors) | Not published |
| Known failure | Text repetition and looping on some pages, hallucinated text reported, and most bold and italic styling is lost | Misses tiny text in large images because of input resolution limits; one user found results on personal documents did not match the promised scores |
Sources
- Hugging Face model card (Unlimited OCR)
- MarkTechPost on Unlimited OCR 2
- Issue #116 OOM on 12 GB
- donut-base model card
- Donut ECCV paper
- Hacker News thread
What each one is
Unlimited OCR
Open-source model from Baidu that reads hundreds of pages in one pass with flat memory use.
Unlimited OCR 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": "baidu/unlimited-ocr","fallbacks": ["naver/donut"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Unlimited OCR alternativesDonut alternativesBest OCR models in 2026