Issues when using Document Intelligence

Question

When using the prebuilt-layout model to extract data from a .pdf file, regardless of how many pages the document contains, the output always includes only two pages. As a result, the remaining pages of the document are missing from the response.

I am using the Azure Document Intelligence REST API (v4.0 GA).

Why is this happening?

Best regards,

Stella

Answer 1

Hello Desenvolvimento nbn,

Thanks for posting on Microsoft Q&A. The prebuilt-layout model (v4.0) processes only the first 2 pages of PDFs on the free tier (F0) due to quota limits; paid tiers (S0+) handle up to 2,000 pages.

​Quick Fixes

• Upgrade tier: Azure portal > Document Intelligence resource > Pricing tier → S0 (5000 transactions/month free, scales) → All pages processed.
• Specify pages (sync API workaround): Add ?pages=1- (all) or ?pages=1-10 to REST endpoint: POST https://{endpoint}/documentintelligence/documentModels/prebuilt-layout:analyze?api-version=2024-11-30&pages=1-.
• ​Use async API (recommended multi-page):
  • POST /documentModels/prebuilt-layout:analyze?api-version=2024-11-30 → Get operationLocation.
  • GET operation → Full multi-page output

Tier 1	PDF Files / FIle Size
Free (F0)	2 / 4MB
Standard (S0)	2000/ 500 MB

Test with Studio (studio.ai.azure.com) to confirm tier/output. Share request/response snippet for more. If it's helpful, please accept the answer.

Best regards,

Jerald Felix