Tesseract vs dots.ocr
Pick Tesseract for clean printed scans that you need to turn into searchable PDFs or plain text on a CPU, in many languages, at low cost. Pick dots.ocr when you need layout, tables as HTML, formulas as LaTeX and Markdown from mixed PDFs, and you have a CUDA GPU. The main split is a CPU line-and-word engine versus a 3B GPU model that parses whole page structure.
Tesseracttesseract/tesseract | dots.ocrrednote/dots-ocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | Free | 2 credits / page |
| Per 1,000 pages | $0.00 | $2.00 |
| Typical speed | 0.6 s / page | 1.5 s / page |
| Best for | Clean printed text, free | Multilingual document parsing |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | 100+ languages |
| Max pages per request | 500 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | No | ✓ 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 Tesseract if
- It runs on CPU only with no model server, so it fits small machines and offline batch jobs.
- It recognizes more than 100 languages out of the box and can be trained for more.
- It outputs plain text, hOCR, searchable PDF, TSV, ALTO and PAGE XML, and it is the engine behind OCRmyPDF.
Choose dots.ocr if
- One pass returns layout boxes, categories and text, with tables as HTML and formulas as LaTeX.
- It outputs JSON, Markdown, HTML, LaTeX and bounding boxes, so mixed academic and business PDFs convert to Markdown with layout.
- The dots.mocr variant reaches 83.9 on olmOCR-bench (authors' figure), up from 79.1 for the original, and dots.mocr-svg converts charts and diagrams to SVG.
Key differences
| Tesseract | dots.ocr | |
|---|---|---|
| Type of tool | LSTM OCR engine, run as a CLI or C++ library | 3B parameter document parsing VLM (1.7B LLM backbone) |
| Licence | Apache-2.0 | MIT for the code; the repository also includes a separate dots.ocr license agreement file |
| Hardware | CPU only, no GPU needed | Single CUDA GPU with vLLM is the documented path; no VRAM figure stated |
| Output | Plain text, hOCR, searchable PDF, TSV, ALTO, PAGE; lines and words only, no table, formula or reading-order output | JSON, Markdown, HTML, LaTeX, bounding boxes, SVG; layout, tables and formulas in one pass |
| Benchmark | Not published | olmOCR-bench 79.1 (dots.ocr) and 83.9 (dots.mocr), both from the authors |
| Input quality | Accuracy drops sharply without binarization, deskew and roughly 300 DPI input; not built for handwriting | Not published |
| Known failure | Confuses digit 1 with lowercase l (issue #2753) | Runs of ellipses or underscores can make output repeat endlessly; a table above 1000 grid cells raised an error and dropped the page in Docling (issue #5146) |
Sources
- Tesseract GitHub README
- Improving quality (tessdoc)
- Issue #2753 1 vs l
- dots.ocr GitHub README
- dots.ocr blog / limitations
- Docling issue #5146
What each one is
dots.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.
PDF, PNG, JPG, WebP, HEIC. Up to 20 MB each, 10 files.No file?
Use one, fall back to the other
{"model": "tesseract/tesseract","fallbacks": ["rednote/dots-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Tesseract vs Nanonets-OCRTesseract vs SuryaPaddleOCR (PP-OCR) vs dots.ocrDocling vs dots.ocrAmazon Textract vs TesseractPaddleOCR-VL vs TesseractDocling vs Tesseractdots.ocr vs Granite Docling
Tesseract alternativesdots.ocr alternativesBest OCR models in 2026