Tesseract vs Typhoon OCR
Pick Tesseract for clean printed scans, offline searchable PDFs and cheap CPU batch jobs in many languages. Pick Typhoon OCR for Thai and English documents with tables, equations and forms, where you need Markdown, HTML or LaTeX output and can run a 2B model. Tesseract only returns lines and words, while Typhoon OCR returns structure, but it supports just two languages.
Tesseracttesseract/tesseract | Typhoon OCRtyphoon/typhoon-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 | 1.8 s / page |
| Best for | Clean printed text, free | Thai documents |
| Inputs | PDF, images | PDF, images |
| Languages | 100+ languages | Thai, English |
| Max pages per request | 500 | 500 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | No |
| Bounding boxes | No | No |
| 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 use.
- It recognizes more than 100 languages out of the box and can be trained for more, while Typhoon OCR was evaluated only on Thai and English.
- It outputs plain text, hOCR, searchable PDF, TSV, ALTO and PAGE XML, and it is the engine behind OCRmyPDF.
Choose Typhoon OCR if
- It returns tables as HTML with merged cells, equations as LaTeX and marked checkboxes, which Tesseract does not produce.
- V1.5 scored BLEU 0.870 on Thai government forms and beat GPT-5 and Gemini 2.5 Pro there.
- V2 2B scores 84.7 on olmOCR-bench at only 2B parameters, and 81.3 on MDPBench Thai pages versus 61.3 for its base model.
Key differences
| Tesseract | Typhoon OCR | |
|---|---|---|
| Licence | Apache-2.0 | Apache-2.0, subject to OpenTyphoon Terms and Conditions |
| How you run it | Local CLI or C++ library on CPU, no GPU needed | Self-hosted 2B weights (0.8B V2 variant also exists), or the Typhoon API as typhoon-ocr-v2 |
| Price | Free, self-hosted (you pay for compute) | Free, self-hosted (you pay for compute). Typhoon API pricing: Not published |
| Output formats | Plain text, hOCR, searchable PDF, TSV, ALTO, PAGE XML | Markdown, HTML, LaTeX |
| Languages | More than 100 out of the box, trainable for more | Thai and English only evaluated |
| Tables and formulas | No table structure, formula or reading-order output; lines and words only | Tables as HTML with merged cells, equations as LaTeX |
| Known failure | Accuracy drops sharply without preprocessing (binarization, deskew, about 300 DPI); confuses digit 1 with lowercase l (issue #2753) | V2 can loop until the token limit on dense multi-column pages; key-information extraction is weakest at 21.0 on ThaiOCRBench |
Sources
- Tesseract GitHub README
- Improving quality (tessdoc)
- Issue #2753 1 vs l
- Typhoon OCR V2 2B model card
- Typhoon OCR technical report
What each one is
Typhoon OCR
Open vision model tuned for Thai and English documents, with Markdown output.
Typhoon 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": ["typhoon/typhoon-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Tesseract vs Florence-2Tesseract vs GOT-OCR 2.0PaddleOCR (PP-OCR) vs Typhoon OCRDocling vs Typhoon OCRAmazon Textract vs TesseractPaddleOCR-VL vs TesseractDocling vs TesseractTyphoon OCR vs OCR.space
Tesseract alternativesTyphoon OCR alternativesBest OCR models in 2026