Amazon Textract vs Nanonets-OCR
Pick Amazon Textract for English forms, invoices and large scanned PDF batches on AWS, where you want JSON with bounding boxes and confidence scores from a managed API. Pick Nanonets-OCR if you want to self-host and turn printed business documents into Markdown for LLMs, including signatures, checkboxes and flowcharts. Avoid Nanonets-OCR for handwriting, and avoid Textract for Chinese, Japanese or other non-Latin text.
Amazon Textractaws/textract | Nanonets-OCRnanonets/nanonets-ocr | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 25 credits / page | 2 credits / page |
| Per 1,000 pages | $25.00 | $2.00 |
| Typical speed | 1.2 s / page | 1.8 s / page |
| Best for | Forms and key-value extraction | Markdown with semantic tags |
| Inputs | PDF, images | PDF, images |
| Languages | 6 languages | Multilingual |
| Max pages per request | 3,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | No |
| Confidence | ✓ Yes | No |
| Forms | ✓ Yes | ✓ Yes |
| Multilingual | No | ✓ 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 Amazon Textract if
- It handles large scanned PDF batches: the async API takes PDFs up to 500 MB and 3,000 pages.
- It returns word and line geometry with confidence scores, so you can check and place text in JSON or bounding boxes.
- Basic text detection costs $0.0015 per page, with Tables at $0.015, Forms at $0.05 and Queries at $0.015 for the first 1M pages a month in US West (Oregon).
Choose Nanonets-OCR if
- It outputs Markdown, HTML, LaTeX and Mermaid, so printed PDFs go to LLMs with no post-processing and flowcharts and org charts become Mermaid code.
- It tags signatures, watermarks and image descriptions, and converts checkboxes to unicode symbols, which suits business documents.
- The weights are on Hugging Face and free to self-host, so you pay only for compute and keep documents on your own hardware.
Key differences
| Amazon Textract | Nanonets-OCR | |
|---|---|---|
| How you run it | Hosted API on AWS | Self-host the weights from Hugging Face |
| Licence | Proprietary | Not stated on the Nanonets-OCR2-3B model card (no license tag); base model is Qwen2.5-VL-3B |
| Price | From $0.0015 per page for text detection; Forms $0.05 per page | Free to self-host; you pay for compute |
| Output formats | JSON, plain text, bounding boxes | Markdown, HTML, LaTeX, Mermaid |
| Benchmark | RD-TableBench table accuracy 80.9% (independent) | olmOCR-bench overall 69.5 and DocVQA 89.43 for OCR2-3B (reported by the authors) |
| Handwriting | Supported, English only | Not trained on handwriting; can hallucinate, and one test saw generation loops with CER up to 2100% |
| Languages and limits | Text detection covers English, French, German, Italian, Portuguese and Spanish; no vertical Japanese or Chinese; sync API takes one page | Fixed prompt; custom extraction prompts are expected to hallucinate |
Sources
- Textract pricing
- Textract quotas
- Textract hard limits
- Reducto RD-TableBench
- Nanonets-OCR2-3B model card
- HF discussion fixed prompt
What each one is
Amazon Textract
AWS OCR with form key-value pairs, tables and per-word confidence scores.
Amazon Textract detailsNanonets-OCR
Open image-to-Markdown model that tags tables, equations, signatures, watermarks and checkboxes.
Nanonets-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": "aws/textract","fallbacks": ["nanonets/nanonets-ocr"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Amazon Textract alternativesNanonets-OCR alternativesBest OCR models in 2026