OCRRouter

Unlimited OCR vs RapidOCR

Pick Unlimited OCR for long, multi-page documents such as reports and books that you want as Markdown with bounding boxes, and you have an NVIDIA GPU. Pick RapidOCR for offline CPU OCR of Chinese and English screenshots and receipts, or when you need to embed OCR in a Java, C# or C++ app. The main split is that Unlimited OCR parses whole documents on a GPU, while RapidOCR reads text lines on a CPU.

Unlimited OCRbaidu/unlimited-ocrRapidOCRrapidai/rapidocr
TypeOpen sourceOpen source
Price2 credits / page1 credit / page
Per 1,000 pages$2.00$1.00
Typical speed1.2 s / page0.6 s / page
Best forBooks and very long documentsLightweight OCR on CPU or edge
InputsPDF, imagesimages
Languages30+ languagesChinese, English and more
Max pages per request1,000100
Tables✓ YesNo
HandwritingNoNo
Math✓ YesNo
Bounding boxes✓ Yes✓ Yes
ConfidenceNo✓ Yes
FormsNoNo
Multilingual✓ Yes✓ Yes
On OCRRouterAvailableAvailable

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.
  • It outputs Markdown, plain text and bounding boxes, with block types alongside the text, which suits layout-aware extraction and PDF to text for RAG.
  • It uses the MIT licence, which allows commercial use with no copyleft, and Baidu Cloud also offers a hosted endpoint.

Choose RapidOCR if

  • It runs on CPU by default through ONNX Runtime, so it works on machines with no GPU.
  • It has bindings for Python, C++, Java and C#, so it embeds in desktop and server apps in many stacks.
  • Install is `pip install rapidocr onnxruntime`, which runs PaddleOCR models without the Paddle framework, and optional engines include TensorRT, OpenVINO and MNN.

Key differences

Unlimited OCRRapidOCR
How you run itSelf-host with vLLM, SGLang or transformers, or use the hosted endpoint in Baidu CloudLocal library from Python, C++, Java or C#; no hosted option listed
HardwareNVIDIA GPU with CUDA 12.9 or 13.0; version 2 under SGLang runs out of memory on 12 GB GPUs; not for CPU-only machinesCPU by default; optional GPU, TensorRT, OpenVINO, MNN, Paddle or PyTorch engines
LicenceMITApache-2.0; converted models keep their upstream licences
Output formatsMarkdown, plain text, bounding boxesPlain text, bounding boxes; not suited to layout, tables or Markdown
Document scopeMulti-page documents in one pass, 40+ pages; context capped at 32,768 tokens and multi-page input only works in base modeNot published
BenchmarksParseBench mean 46.17 (independent); Unlimited OCR 2 scores 84.6 on MPDocBench and 77.1 on MarathonDocBench (by the authors)Not published
LanguagesNot publishedChinese and English only by default; other languages need extra models
Known failuresText repetition and looping (issue #55), hallucinated text (issue #58), bbox coordinate loops on GB10 GPUs under SGLang (issue #114); bold and italic styling mostly lost16-bit grayscale PNG or TIFF give no text or wrong text (issue #776); sliced inference crashes on large pages (issue #764); Arabic crashes (issue #719)

Sources

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 details

RapidOCR

PaddleOCR models packaged for ONNX Runtime and other engines. Easy to deploy without Paddle.

RapidOCR details

Try both on your own file

Run the file once with each model and compare the output.

PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?

Use one, fall back to the other

{"model": "baidu/unlimited-ocr","fallbacks": ["rapidai/rapidocr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Unlimited OCR alternativesRapidOCR alternativesBest OCR models in 2026