Amazon Textract vs Florence-2
Pick Amazon Textract for English forms, invoices and large scanned PDF batches on AWS, where you need tables, forms and word-level confidence scores. Pick Florence-2 for short labels and scene text in English, where you want a small MIT-licensed model you host yourself and get boxes per text line. Neither suits Chinese or Japanese text.
Amazon Textractaws/textract | Florence-2microsoft/florence-2 | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 25 credits / page | 1 credit / page |
| Per 1,000 pages | $25.00 | $1.00 |
| Typical speed | 1.2 s / page | 0.8 s / page |
| Best for | Forms and key-value extraction | Lightweight OCR with boxes |
| Inputs | PDF, images | images |
| Languages | 6 languages | English |
| Max pages per request | 3,000 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | ✓ Yes | No |
| Math | No | No |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | ✓ Yes | No |
| Multilingual | No | 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 Amazon Textract if
- It extracts tables, forms, queries and layout, and Florence-2 returns only text and boxes.
- The async API takes PDFs up to 500 MB and 3,000 pages, which fits large scanned batches.
- It returns word and line geometry with confidence scores, and scored 80.9% on RD-TableBench in an independent test.
Choose Florence-2 if
- It is MIT licensed and free to self-host, and you pay only for compute.
- It is small, at 0.23B (base) and 0.77B (large) parameters, so it suits lightweight edge OCR in English.
- OCR_WITH_REGION returns quadrilateral boxes per text line, and the same model also does captioning, detection and grounding.
Key differences
| Amazon Textract | Florence-2 | |
|---|---|---|
| How you run it | Hosted AWS API | Self-hosted via Transformers |
| Licence | Proprietary | MIT |
| 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 | JSON, plain text, bounding boxes | Plain text, bounding boxes. No Markdown, tables or LaTeX |
| Document structure | Tables, forms, queries and layout | None. Text and boxes only |
| Languages | Text detection: English, French, German, Italian, Portuguese, Spanish. Handwriting and Queries: English only. Vertical Japanese and Chinese not supported | Chinese output is mostly wrong because the vocabulary lacks CJK characters. Spelling errors and missing spaces seen on French text |
| Handwriting | Supported, English only | Errors and missing spaces on cursive handwriting in one blog test |
Sources
- Textract pricing
- Textract quotas
- Reducto RD-TableBench
- Florence-2-large model card
- HF discussion Chinese very bad
What each one is
Amazon Textract
AWS OCR with form key-value pairs, tables and per-word confidence scores.
Amazon Textract detailsFlorence-2
Small Microsoft vision model (0.23B and 0.77B) with OCR and OCR-with-regions tasks that return text with boxes.
Florence-2 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": ["microsoft/florence-2"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Amazon Textract alternativesFlorence-2 alternativesBest OCR models in 2026