Azure Document Intelligence vs Florence-2
Pick Azure Document Intelligence for invoices, receipts, IDs and other structured business documents, especially if you already run on Azure. Pick Florence-2 for free, self-hosted OCR of English scene text and short labels with boxes. The main split is scope: Azure returns document structure (tables, Markdown, prebuilt fields), while Florence-2 returns only text and boxes.
Azure Document Intelligenceazure/document-intelligence | Florence-2microsoft/florence-2 | |
|---|---|---|
| Type | OCR API | Open source |
| Price | 60 credits / page | 1 credit / page |
| Per 1,000 pages | $60.00 | $1.00 |
| Typical speed | 1.5 s / page | 0.8 s / page |
| Best for | Enterprise forms and IDs | Lightweight OCR with boxes |
| Inputs | PDF, images | images |
| Languages | 300+ languages | English |
| Max pages per request | 2,000 | 500 |
| Tables | ✓ Yes | No |
| Handwriting | ✓ Yes | No |
| Math | No | No |
| Bounding boxes | ✓ Yes | ✓ Yes |
| Confidence | ✓ Yes | No |
| Forms | ✓ Yes | 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 Azure Document Intelligence if
- It has prebuilt models for invoices, receipts, IDs, W-2 and contracts, so you get fields without training anything.
- It outputs JSON, Markdown, plain text, bounding boxes and searchable PDF, so full-page documents and tables are covered.
- It runs as a managed service or in disconnected containers for on-prem use, and Read OCR costs $1.50 per 1,000 pages.
Choose Florence-2 if
- It is MIT licensed and free, and you only pay for the compute you run it on.
- At 0.23B (base) and 0.77B (large) parameters, it is small enough for 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
| Azure Document Intelligence | Florence-2 | |
|---|---|---|
| Licence | Proprietary | MIT |
| How you run it | Managed service or containers, including disconnected on-prem | Self-host via Transformers |
| Price | Read $1.50 per 1,000 pages ($0.60 above 1M); Layout and prebuilt $10 per 1,000; custom extraction $30 per 1,000; 500 free pages per month | Free; you pay for compute |
| Output formats | JSON, Markdown, plain text, bounding boxes, searchable PDF | Plain text and bounding boxes only; no Markdown, tables or LaTeX |
| Tables | 82.7% on RD-TableBench (independent test); Layout is $10 per 1,000 pages | No table output |
| Handwriting and non-English text | Not published | Spelling errors and missing spaces on cursive handwriting and French text; Chinese output mostly wrong because the vocabulary lacks CJK characters |
| Known failure | Stacked models multiply cost: Prebuilt plus Layout plus high-res add-on is $26 per 1,000 pages; the free tier reads only the first two pages of each document | Model card notes the continued-pretrained checkpoint used only 0.1B samples and might not be well trained |
Sources
- Azure pricing page
- Azure DI pricing breakdown
- Reducto RD-TableBench
- Florence-2-large model card
- HF blog OCR test
- HF discussion Chinese very bad
What each one is
Azure Document Intelligence
Azure layout model with tables, selection marks and prebuilt invoice and ID readers.
Azure Document Intelligence 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": "azure/document-intelligence","fallbacks": ["microsoft/florence-2"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Azure Document Intelligence alternativesFlorence-2 alternativesBest OCR models in 2026