PaddleOCR-VL vs Typhoon OCR
Pick PaddleOCR-VL for multilingual documents (109 languages) with formulas, tables and charts, run on your own GPU. Pick Typhoon OCR for Thai and English documents such as government forms and financial reports, or when you want a hosted API option. The main split is language coverage: PaddleOCR-VL is broad, while Typhoon OCR is tuned for Thai and only Thai and English were evaluated.
PaddleOCR-VLpaddle/paddleocr-vl | Typhoon OCRtyphoon/typhoon-ocr | |
|---|---|---|
| Type | Open source | Open source |
| Price | 6 credits / page | 2 credits / page |
| Per 1,000 pages | $6.00 | $2.00 |
| Typical speed | 1.3 s / page | 1.8 s / page |
| Best for | CJK documents | Thai documents |
| Inputs | PDF, images | PDF, images |
| Languages | 109 languages | Thai, English |
| Max pages per request | 500 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | No | No |
| Math | ✓ Yes | No |
| Bounding boxes | ✓ Yes | No |
| Confidence | ✓ Yes | 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 PaddleOCR-VL if
- It supports 109 languages in one model, including Cyrillic, Arabic, Devanagari and Thai scripts.
- It recognizes text, tables, formulas and charts, and outputs Markdown and JSON for full pages.
- At under 1B parameters it runs on much smaller GPUs than 7B document VLMs like olmOCR.
Choose Typhoon OCR if
- It beat GPT-5 and Gemini 2.5 Pro on Thai government forms in V1.5 (BLEU 0.870), and V2 2B scores 81.3 on MDPBench Thai pages versus 61.3 for its base model.
- It outputs tables as HTML with merged cells, equations as LaTeX, and marks checkboxes.
- You can self-host the weights or call it on the Typhoon API as typhoon-ocr-v2.
Key differences
| PaddleOCR-VL | Typhoon OCR | |
|---|---|---|
| Languages | 109 languages, including Cyrillic, Arabic, Devanagari and Thai | Thai and English only evaluated |
| How you run it | Self-host only, via PaddleOCR, vLLM or transformers | Self-host the weights or use the Typhoon API (typhoon-ocr-v2) |
| Licence | Apache-2.0 | Apache-2.0, subject to OpenTyphoon Terms and Conditions |
| Size | 0.9B parameters | 2B parameters (V2 and V1.5); 0.8B V2 variant |
| Output formats | Markdown, JSON | Markdown, HTML, LaTeX |
| olmOCR-Bench | 80.0 (independent) | 84.7 (V2 2B, reported by the authors) |
| Known failure | Version 1.6 hangs on blurry images, and decoding is about 11x slower because use_cache ships as false. It also regresses Russian and Kazakh OCR on born-digital PDFs versus 1.5. | V2 can loop until the token limit, mostly on dense multi-column pages. Key-information extraction is weakest, at 21.0 on ThaiOCRBench. |
Sources
- PaddleOCR-VL-1.6 model card
- olmOCR-Bench table
- Issue #18453 use_cache slowdown
- Typhoon OCR V2 2B model card
- Typhoon OCR technical report
What each one is
PaddleOCR-VL
Open-source document model from Baidu. Excellent on Chinese, Japanese and Korean.
PaddleOCR-VL detailsTyphoon 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.
Use one, fall back to the other
{"model": "paddle/paddleocr-vl","fallbacks": ["typhoon/typhoon-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
PaddleOCR-VL alternativesTyphoon OCR alternativesBest OCR models in 2026