Azure Document Intelligence vs InternVL3.5
Pick InternVL3.5 if you can run GPUs and want document question answering or Markdown/JSON output under an Apache-2.0 licence, with no per-page fee. Pick Azure Document Intelligence if you process invoices, receipts or IDs, or need word coordinates, and want a managed service with prebuilt models. The main split is how you run it: Azure bills per page and returns layout data, while InternVL3.5 is free weights you host and size yourself.
Azure Document Intelligenceazure/document-intelligence | InternVL3.5opengvlab/internvl | |
|---|---|---|
| Type | OCR API | Vision model |
| Price | 60 credits / page | 2 credits / page |
| Per 1,000 pages | $60.00 | $2.00 |
| Typical speed | 1.5 s / page | 3.0 s / page |
| Best for | Enterprise forms and IDs | Self-hosted document understanding |
| Inputs | PDF, images | images |
| Languages | 300+ languages | Multilingual |
| Max pages per request | 2,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | No |
| Math | No | ✓ Yes |
| Bounding boxes | ✓ Yes | No |
| Confidence | ✓ Yes | No |
| Forms | ✓ Yes | No |
| Multilingual | ✓ Yes | ✓ 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 Azure Document Intelligence if
- Prebuilt models cover invoices, receipts, IDs, W-2 and contracts, so you skip training and prompt work for those document types.
- It returns bounding boxes and searchable PDF alongside JSON, Markdown and plain text, which suits jobs that need word positions.
- It is a managed service with disconnected containers for on-prem use, so you do not need to run your own GPUs.
Choose InternVL3.5 if
- Weights are Apache-2.0 and free, so you pay only for compute and no per-page charge applies.
- Nine sizes from 1B to 241B let you trade accuracy for cost, and models up to 30B fit on a single A100.
- It suits document question answering and self-hosted multimodal pipelines, where one model family covers several sizes.
Key differences
| Azure Document Intelligence | InternVL3.5 | |
|---|---|---|
| How you run it | Managed Azure service; also available as containers, including disconnected ones | Self-host only, with Transformers, LMDeploy or vLLM |
| Licence | Proprietary | Apache-2.0 |
| Price | Read $1.50 per 1,000 pages; Layout and prebuilt $10 per 1,000; custom extraction $30 per 1,000; 500 free pages a month | Free; you pay for compute |
| Output formats | JSON, Markdown, plain text, bounding boxes, searchable PDF | Plain text, Markdown, JSON; no bounding boxes or layout output by default |
| Table benchmark | 82.7% on RD-TableBench (independent) | Not published |
| Hardware | None to manage on the hosted service | Up to 30B on one A100; 38B needs two A100s; 241B-A28B needs eight A100s |
| Known weakness | Stacked models add up: Prebuilt plus Layout plus high-res add-on is $26 per 1,000 pages; free tier covers only the first two pages of each document | OCR scores on the model card are chart images, not text values; not suited to CPU-only use of larger sizes |
Sources
- Azure pricing page
- Azure DI pricing breakdown
- Reducto RD-TableBench
- Hugging Face model card (8B)
- InternVL GitHub
What each one is
Azure Document Intelligence
Azure layout model with tables, selection marks and prebuilt invoice and ID readers.
Azure Document Intelligence detailsInternVL3.5
Shanghai AI Lab's open vision-language family (1B to 241B, August 2025), strong on OCR and document benchmarks.
InternVL3.5 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": ["opengvlab/internvl"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Azure Document Intelligence alternativesInternVL3.5 alternativesBest OCR models in 2026