Qianfan OCR vs docTR
DocTR is faster; only Qianfan OCR supports tables, math, handwriting, multilingual; only docTR supports bounding boxes, confidence.
Qianfan OCRbaidu/qianfan-ocr | docTRmindee/doctr | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 1 credit / page | 1 credit / page |
| Per 1,000 pages | $1.00 | $1.00 |
| Typical speed | 2.0 s / page | 0.9 s / page |
| Best for | Chinese documents | Word-level OCR with positions |
| Inputs | PDF, images | PDF, images |
| Languages | Chinese, English and more | Mainly Latin scripts |
| Max pages per request | 200 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | ✓ Yes | No |
| Math | ✓ Yes | No |
| Bounding boxes | No | ✓ Yes |
| Confidence | No | ✓ Yes |
| Forms | No | No |
| Multilingual | ✓ Yes | No |
| On OCRRouter | Available | Coming soon |
Prices are what you pay on OCRRouter. Speed is the provider's typical figure, not a benchmark we ran.
What each one is
Qianfan OCR
Baidu's OCR-tuned vision model. Strong on Chinese documents and tables.
Qianfan OCR detailsdocTR
Open-source OCR library from Mindee with separate text detection and recognition models.
docTR detailsUse one, fall back to the other
Request body
{
"model": "baidu/qianfan-ocr",
"fallbacks": ["mindee/doctr"],
"file_url": "https://example.com/document.pdf"
}