Unlimited OCR vs dots.ocr
Pick Unlimited OCR for long reports and books that you want parsed in one request, such as bulk PDF to text for RAG. Pick dots.ocr for multilingual business and academic documents where you need tables as HTML, formulas as LaTeX, or chart-to-SVG output. The main split is document length versus output richness: Unlimited OCR handles 40+ pages per pass, dots.ocr returns more structured formats per page.
Unlimited OCRbaidu/unlimited-ocr | dots.ocrrednote/dots-ocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 2 credits / page | 2 credits / page |
| Per 1,000 pages | $2.00 | $2.00 |
| Typical speed | 1.2 s / page | 1.5 s / page |
| Best for | Books and very long documents | Multilingual document parsing |
| Inputs | PDF, images | PDF, images |
| Languages | 30+ languages | 100+ languages |
| Max pages per request | 1,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | No | No |
| 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 Unlimited OCR if
- It parses 40+ pages in a single forward pass with roughly constant memory, so long reports and books go in as one request.
- Baidu Cloud offers a hosted endpoint, so you can skip running GPUs yourself if you prefer.
- Unlimited OCR 2 scores 77.1 on MarathonDocBench against 33.7 for the next model, per Baidu.
Choose dots.ocr if
- It outputs JSON, Markdown, HTML, LaTeX and SVG, with tables as HTML and formulas as LaTeX.
- The dots.mocr-svg variant converts charts and diagrams to SVG.
- dots.mocr (v1.5) reaches 83.9 on olmOCR-bench, up from 79.1 for the original, and it is built as a multilingual parser.
Key differences
| Unlimited OCR | dots.ocr | |
|---|---|---|
| How you run it | Self-host with vLLM, SGLang or transformers, or use the hosted endpoint in Baidu Cloud's OCR product | Self-host with vLLM or Hugging Face Transformers; no hosted option listed |
| Licence | MIT | MIT, with a separate dots.ocr license agreement file in the repository |
| Output formats | Markdown, plain text, bounding boxes | JSON, Markdown, HTML, LaTeX, bounding boxes, SVG |
| Long documents | 40+ pages in one forward pass; context capped at 32,768 tokens, and multi-page input only works in base mode | Not published |
| Hardware | NVIDIA GPU with CUDA 12.9 or 13.0; version 2 under SGLang runs out of memory on 12 GB GPUs | Single CUDA GPU with vLLM; no VRAM figure stated |
| Text formatting | ParseBench text formatting score is 0.97, so bold and italic styling is mostly lost | Not published |
| Known failures | Text repetition and looping, hallucinated text, and bbox coordinate loops on GB10 GPUs with SGLang | Endless repeats on runs of ellipses or underscores; tables above 1000 grid cells raise an error and drop the page in Docling; complex tables and formulas remain difficult |
Sources
- Unlimited OCR Hugging Face model card
- Unlimited OCR GitHub repo
- MarkTechPost on Unlimited OCR 2
- dots.ocr GitHub README
- Docling issue #5146
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 detailsdots.ocr
Single vision model from rednote that does layout detection and content recognition together.
dots.ocr 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": ["rednote/dots-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Unlimited OCR alternativesdots.ocr alternativesBest OCR models in 2026