OCRRouter

SmolDocling vs QVQ-72B-Preview

SmolDocling is cheaper per page. SmolDocling is faster. Only SmolDocling supports Tables, Math, Bounding boxes. Only QVQ-72B-Preview supports Multilingual.

SmolDoclinghuggingface/smoldoclingQVQ-72B-Previewqwen/qvq-72b
TypeOpen sourceOpen source
Price1 credit / page3 credits / page
Per 1,000 pages$1.00$3.00
Typical speed1.0 s / page3.0 s / page
Best forVery cheap page conversionVisual reasoning over charts
InputsPDF, imagesimages
LanguagesMainly EnglishMainly English and Chinese
Max pages per request50050
Tables✓ YesNo
HandwritingNoNo
Math✓ YesNo
Bounding boxes✓ YesNo
ConfidenceNoNo
FormsNoNo
MultilingualNo✓ Yes
On OCRRouterAvailableAvailable

Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.

What each one is

SmolDocling

Tiny 256M vision model from Hugging Face and IBM that outputs DocTags. Predecessor of Granite Docling.

SmolDocling details

QVQ-72B-Preview

Qwen's experimental visual reasoning model (December 2024).

QVQ-72B-Preview 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 10 files at onceNo file?

Use one, fall back to the other

{"model": "huggingface/smoldocling","fallbacks": ["qwen/qvq-72b"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

SmolDocling alternativesQVQ-72B-Preview alternativesBest OCR models in 2026