Amazon Textract vs MinerU
Pick Amazon Textract for English forms, invoices and large scanned PDF batches when you already run on AWS and want a managed API with per-page pricing. Pick MinerU for Chinese and English academic papers or mixed Office and PDF corpora where you need Markdown out and can host it yourself. The main split is a hosted, proprietary API limited to six Latin-script languages versus a self-hosted toolkit with a custom licence.
Amazon Textractaws/textract | MinerUopendatalab/mineru | |
|---|---|---|
| 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 | 2.0 s / page |
| Best for | Forms and key-value extraction | Academic PDFs |
| Inputs | PDF, images | PDF, images |
| Languages | 6 languages | 80+ languages |
| Max pages per request | 3,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | ✓ Yes | No |
| 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
- You need English forms, invoices, handwriting or Queries, and you want them from a managed AWS API.
- You process big scanned PDF batches: the async API takes files up to 500 MB and 3,000 pages.
- You want predictable per-page pricing, from $0.0015 per page for text detection up to $0.05 per page for Forms, with word and line geometry and confidence scores in the output.
Choose MinerU if
- You work with Chinese and English academic papers, or a mixed corpus of PDF, DOCX, PPTX, XLSX, EPUB, OFD and HTML files for RAG.
- You need Markdown, HTML, LaTeX, DOCX, EPUB or JSON output without post-processing.
- You want to run on your own hardware, including a CPU-only basic tier with about 2 GB RAM, and pay only for compute.
Key differences
| Amazon Textract | MinerU | |
|---|---|---|
| How you run it | Hosted AWS API | Self-hosted as a CLI, API or WebUI |
| Licence | Proprietary | MinerU Open Source License (Apache-2.0 based with extra conditions); MinerU2.5 VLM weights are tagged AGPL-3.0 |
| Price | US West (Oregon), first 1M pages/month: text detection $0.0015/page, Tables $0.015/page, Forms $0.05/page, Queries $0.015/page | Free; you pay for compute |
| Output formats | JSON, plain text, bounding boxes; no Markdown without post-processing | Markdown, HTML, LaTeX, DOCX, EPUB, JSON |
| Languages | Text detection supports English, French, German, Italian, Portuguese and Spanish; handwriting and Queries are English only; vertical Japanese and Chinese not supported | Listed as good for Chinese and English academic papers; full-width Chinese punctuation is converted to half-width with no switch |
| Benchmark | RD-TableBench table accuracy: 80.9% | olmOCR-Bench overall: 75.2 (MinerU 2.5.4); ParseBench mean: 45.9 (MinerU2.5). Different benchmarks, so not directly comparable |
| Known failure | Sync API takes one page only | VLM tiers drop merged header rows in tables such as bank statement account lines; the VLM pipeline can produce repeated content, such as 300 KB figure blocks |
Sources
- Textract pricing
- Textract quotas
- Textract hard limits
- Reducto RD-TableBench
- MinerU GitHub README
- Issue #5554 table header rows
What each one is
Amazon Textract
AWS OCR with form key-value pairs, tables and per-word confidence scores.
Amazon Textract detailsMinerU
Open-source PDF to Markdown pipeline popular for academic and Chinese documents.
MinerU 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": ["opendatalab/mineru"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Amazon Textract alternativesMinerU alternativesBest OCR models in 2026