OCRRouter

Hunyuan OCR vs MonkeyOCR

Pick MonkeyOCR for English and Simplified Chinese PDFs, especially academic papers with formulas and tables, as long as you stay on research or buy a commercial licence. Pick Hunyuan OCR for photographed text, translation, and low-resource or historical scripts, or when you need to run on a laptop. The main split is input type: MonkeyOCR is built for clean documents, Hunyuan OCR covers scene text and more task types.

Hunyuan OCRtencent/hunyuan-ocrMonkeyOCRmonkeyocr/monkeyocr
TypeOpen sourceOpen source
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed1.5 s / page1.5 s / page
Best forChinese documents and photosChinese and English PDFs
InputsPDF, imagesPDF, images
Languages100+ languagesChinese, English
Max pages per request500500
Tables✓ Yes✓ Yes
HandwritingNoNo
Math✓ Yes✓ Yes
Bounding boxes✓ Yes✓ Yes
ConfidenceNoNo
FormsNoNo
Multilingual✓ YesNo
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 Hunyuan OCR if

  • It handles photo and scene text spotting, which MonkeyOCR does not fully support.
  • One roughly 1B model covers parsing, spotting, information extraction and image translation, with 12 task types in the vLLM client including low-resource and ancient-script OCR.
  • It runs on a CPU, consumer GPU or laptop through the llama.cpp GGUF path, and v1.5 accepts images up to 4K with a 128K context window.

Choose MonkeyOCR if

  • It targets English and Simplified Chinese PDFs, and its pro-3B model reports OmniDocBench overall edit distance of 0.138 (EN) and 0.206 (ZH) and table TEDS of 81.5 (EN) and 87.5 (ZH).
  • The 1.2B model is about 36% faster than pro-3B for about 1.6% lower accuracy, and quantized weights run on an 8GB RTX 4060.
  • Its code is Apache-2.0, and output includes Markdown, JSON with block coordinates and a layout PDF.

Key differences

Hunyuan OCRMonkeyOCR
LicenceTencent Hunyuan Community License. Excludes the EU, UK and South Korea. Separate licence needed above 100M monthly active users.Code is Apache-2.0. v1 weights are research and non-commercial only. Commercial use or SaaS needs a written licence.
Model sizeAbout 1B parameters.3B and 1.2B.
How you run itSelf-hosted via vLLM, Transformers or llama.cpp. CUDA 13 and vLLM 0.25.1+ recommended for the unified environment.Self-hosted via LMDeploy or Docker. Docker on 20/30/40-series and V100 needs a patched image to avoid a shared-memory error.
HardwareNVIDIA GPU with CUDA 13, or llama.cpp GGUF on CPU, consumer GPU or laptop.Tested on RTX 3090/4090, A6000, H800 and A100. RTX 4060 8GB works with quantized models.
Output formatsMarkdown, HTML, LaTeX, JSON, plain text, bounding boxes.Markdown, JSON, LaTeX, HTML, bounding boxes.
BenchmarkParseBench-2 overall 67.11 and ParseBench mean 63.86 (v1.5, independent).OmniDocBench overall edit distance 0.138 EN / 0.206 ZH, table TEDS 81.5 EN / 87.5 ZH (pro-3B, reported by the authors). The benchmarks differ, so the scores are not directly comparable.
Known weak spotsArchived 1.0 note said Transformers and vLLM outputs differed in accuracy. ParseBench-2 text score is 62.88.Photographed text, handwriting, Traditional Chinese and multilingual text are not fully supported. Pro-3B runs at about 0.49 pages/s on a 3090.

Sources

What each one is

Hunyuan OCR

Small end-to-end OCR model from Tencent covering parsing, extraction and photo translation.

Hunyuan OCR details

MonkeyOCR

Open-source document parser that splits work into structure, recognition and relation steps.

MonkeyOCR 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": "tencent/hunyuan-ocr","fallbacks": ["monkeyocr/monkeyocr"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

Hunyuan OCR alternativesMonkeyOCR alternativesBest OCR models in 2026