OCRRouter

dots.ocr vs Google Cloud Vision OCR

Pick dots.ocr if you need Markdown, HTML tables and LaTeX formulas from PDFs and can run a GPU yourself. Pick Google Cloud Vision OCR if you need plain text from photos, labels or handwriting at volume with no servers to run. The main split is structure: dots.ocr returns layout and tables, Cloud Vision returns text and boxes only.

dots.ocrrednote/dots-ocrGoogle Cloud Vision OCRgoogle/cloud-vision
TypeOpen sourceOCR API
Price2 credits / page2 credits / page
Per 1,000 pages$2.00$2.00
Typical speed1.5 s / page1.0 s / page
Best forMultilingual document parsingPhotos and simple scans
InputsPDF, imagesPDF, images
Languages100+ languages100+ languages
Max pages per request5002,000
Tables✓ YesNo
HandwritingNo✓ Yes
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 dots.ocr if

  • You need Markdown, HTML tables and LaTeX formulas from multilingual PDFs, such as academic and business documents with mixed tables and text.
  • You want MIT-licensed code you host yourself, with no per-page fee beyond your own compute.
  • You want one model that returns layout boxes, categories and text in a single pass, with a dots.mocr-svg variant that converts charts to SVG.

Choose Google Cloud Vision OCR if

  • You need plain text from photos of signs and labels, or from handwritten notes via DOCUMENT_TEXT_DETECTION with language hints.
  • You want pay-per-use pricing: the first 1,000 units per month are free, then $1.50 per 1,000, dropping to $0.60 per 1,000 above 5M units.
  • You want a hosted API with page, block, paragraph, word and symbol hierarchy and bounding polygons, and async batches of up to 2,000 image files per request.

Key differences

dots.ocrGoogle Cloud Vision OCR
How you run itSelf-hosted via vLLM or Hugging Face Transformers. Single CUDA GPU with vLLM is the documented path.Hosted API on Google Cloud, called over REST or client libraries. No on-premise option.
LicenceMIT for the code, with a separate dots.ocr licence agreement file in the repository.Proprietary.
PriceFree software. You pay for compute.First 1,000 units per month free per feature, then $1.50 per 1,000 up to 5M, $0.60 per 1,000 above. Each PDF page is one unit.
Output formatsJSON, Markdown, HTML, LaTeX, bounding boxes, SVG.JSON, plain text, bounding boxes. No Markdown.
Tables and layoutReturns layout boxes and categories, with tables as HTML. Complex tables remain difficult, and a table above 1000 grid cells raised an error and dropped the whole page in a Docling issue.No table or layout structure. Google points to Document AI for forms and tables.
BenchmarksolmOCR-bench: 79.1 for dots.ocr, 83.9 for dots.mocr (v1.5), as reported by the authors.Not published
Known failureRuns of ellipses or underscores can make output repeat endlessly. Occasional parsing failures still occur.You rebuild reading order yourself. PDF and TIFF need async processing via Cloud Storage, and billing must be enabled even for the free units.

Sources

What each one is

dots.ocr

Single vision model from rednote that does layout detection and content recognition together.

dots.ocr details

Google Cloud Vision OCR

Google Cloud's image OCR with text and document text detection, word boxes and confidence.

Google Cloud Vision OCR 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": "rednote/dots-ocr","fallbacks": ["google/cloud-vision"],"file_url": "https://example.com/document.pdf"}

Questions

More comparisons

dots.ocr alternativesGoogle Cloud Vision OCR alternativesBest OCR models in 2026