Azure Document Intelligence vs MiniCPM-V 4.5
Pick Azure Document Intelligence for invoices, receipts and ID documents when you want a managed API, per-page pricing and prebuilt models, or when you already run on Azure. Pick MiniCPM-V 4.5 when you want Apache-2.0 weights you run yourself on one GPU or an edge device and can live without bounding boxes. The main split is a hosted, priced service with layout output versus a free, self-hosted vision language model that returns text, Markdown or JSON only.
Azure Document Intelligenceazure/document-intelligence | MiniCPM-V 4.5openbmb/minicpm-v | |
|---|---|---|
| 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 | 2.5 s / page |
| Best for | Enterprise forms and IDs | OCR on edge devices |
| Inputs | PDF, images | images |
| Languages | 300+ languages | 30+ languages |
| Max pages per request | 2,000 | 500 |
| Tables | ✓ Yes | ✓ Yes |
| Handwriting | ✓ Yes | ✓ Yes |
| Math | No | No |
| 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
- You need bounding boxes, a searchable PDF, or JSON, Markdown and plain text output, and MiniCPM-V 4.5 gives no coordinates.
- You process invoices, receipts, IDs, W-2s or contracts and want prebuilt models for them.
- You want Read OCR at $1.50 per 1,000 pages ($0.60 above 1M), with disconnected containers for on-prem use.
Choose MiniCPM-V 4.5 if
- You want Apache-2.0 weights and code, with no per-page fee and only your own compute to pay for.
- You need on-device or edge OCR, using int4, GGUF or AWQ builds on llama.cpp or vLLM.
- You want local document Q&A on a single GPU, as a general vision language model that answers questions about a page.
Key differences
| Azure Document Intelligence | MiniCPM-V 4.5 | |
|---|---|---|
| Licence | Proprietary | Apache-2.0 for weights and code |
| How you run it | Managed Microsoft service, also available as containers (disconnected containers for on-prem) | Self-host only, with llama.cpp, vLLM, Ollama or 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, Markdown, JSON; no bounding box or layout output |
| Table benchmark | 82.7% on RD-TableBench (independent) | Not published; the model card shows scores only as images |
| Model size | Not published | 8.7B parameters, built on Qwen3-8B and SigLIP2-400M |
| Known weak spot | Stacked models multiply cost, e.g. 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 | Ollama and SGLang support still pending official releases on the model card; no int4 VRAM figure published |
Sources
What each one is
Azure Document Intelligence
Azure layout model with tables, selection marks and prebuilt invoice and ID readers.
Azure Document Intelligence detailsMiniCPM-V 4.5
8B vision model from OpenBMB (August 2025) built to run on devices, with strong OCR and document parsing.
MiniCPM-V 4.5 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": ["openbmb/minicpm-v"],"file_url": "https://example.com/document.pdf"}Questions
More comparisons
Azure Document Intelligence alternativesMiniCPM-V 4.5 alternativesBest OCR models in 2026