PaddleOCR (PP-OCR) vs Nougat
Pick PaddleOCR (PP-OCR) for printed text lines, receipts, signs and screenshots, especially Chinese and English, on CPU, edge or in the browser, and you can use it commercially under Apache-2.0. Pick Nougat only for English scientific papers where you need Markdown with LaTeX math and tables, and only for non-commercial work. The main split is the job: PaddleOCR returns text lines and boxes, Nougat turns whole paper pages into Markdown.
PaddleOCR (PP-OCR)paddle/paddleocr | Nougatmeta/nougat | |
|---|---|---|
| Type | Open source | Open source |
| Price | 1 credit / page | 2 credits / page |
| Per 1,000 pages | $1.00 | $2.00 |
| Typical speed | 0.7 s / page | 3.0 s / page |
| Best for | Fast printed text, Chinese and English | arXiv-style papers |
| Inputs | PDF, images | PDF, images |
| Languages | 80+ languages | Mainly English |
| Max pages per request | 500 | 200 |
| Tables | No | ✓ Yes |
| Handwriting | No | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | No |
| Confidence | ✓ Yes | No |
| Forms | No | No |
| Multilingual | ✓ Yes | No |
| 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 (PP-OCR) if
- It is Apache-2.0, so you can ship it in a commercial product, while Nougat's weights are CC-BY-NC.
- One PP-OCRv6 model covers 50 languages, including Chinese, English, Japanese and 46 Latin-script languages.
- PP-OCRv6 medium is 34.5M parameters and runs on CPU, NVIDIA GPU and other accelerators, with a 5.2x CPU speedup via OpenVINO, so it fits edge and mobile use.
Choose Nougat if
- It converts scientific PDF pages into Mathpix-style Markdown with LaTeX for equations and tables, which PaddleOCR does not do on its own.
- It was trained on arXiv and PubMed Central documents, so it suits English academic papers and math-heavy PDFs.
- It ships small and base checkpoints with a simple CLI and API, so a paper-to-Markdown job is easy to set up.
Key differences
| PaddleOCR (PP-OCR) | Nougat | |
|---|---|---|
| Licence | Apache-2.0 | Code MIT; weights CC-BY-NC (no commercial use) |
| How you run it | Local library: paddleocr Python package, C++, or the PaddleOCR.js browser SDK | Self-hosted Python package with CLI and API |
| Output | Plain text, bounding boxes, JSON | Markdown and LaTeX (.mmd) |
| Tables and formulas | Not handled by PP-OCR; needs the separate PP-Structure or VL pipelines | LaTeX for equations and tables, built in |
| Languages | 50 languages in one PP-OCRv6 model, including Chinese and Japanese | Best on English; README says Chinese, Russian and Japanese do not work |
| Hardware | CPU, NVIDIA GPU, Kunlunxin XPU; 0.13s on A100 reported for PP-OCRv6 | Not published |
| Known failure | PP-OCRv6 weights contain denormals that slow CPU inference 4 to 11x (RapidOCR issue #743) | Repetition loops on 1.5% of test pages in the paper, more on out-of-domain documents; repetition detection can drop pages as MISSING_PAGE |
Sources
- PaddleOCR GitHub README
- RapidOCR issue #743 denormal weights
- Nougat GitHub README
- Nougat paper
- Issue #11 repetitions
What each one is
PaddleOCR (PP-OCR)
Baidu's classic open-source text detection and recognition pipeline. Light enough to run on CPU.
PaddleOCR (PP-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","fallbacks": ["meta/nougat"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
PaddleOCR (PP-OCR) alternativesNougat alternativesBest OCR models in 2026