OCRRouter

Azure Document Intelligence API docs

How OCRRouter calls Azure Document Intelligence, what the upstream returns field by field, and how that becomes Markdown and JSON blocks.

Overview

Azure Document Intelligence prebuilt-layout model, reached through OpenParser. It returns paragraphs with roles (title, section heading, header, footer), tables, selection marks and per-word confidence.

OCRRouter turns OpenParser's openparser@1 output into per-page Markdown plus layout blocks with boxes in pixels (images) or the page unit Azure reports.

Price and limits

Model ID
azure/document-intelligence
Price
60 credits / page ($60 per 1,000 pages)
Max pages per request
2,000
Inputs
PDF; Images (JPEG, PNG, BMP, TIFF, HEIF per Azure)
Languages
300+ languages (Azure Read/Layout language list)
Positions (bbox)
Yes, when the upstream returns them
  • Azure: up to 2,000 pages per PDF or TIFF, 500 MB file on paid tier
  • OCRRouter catalog limit: 2,000 pages

You pay only for pages that succeed. A failed attempt that falls back to another model is not charged. See pricing.

How OCRRouter calls it

Endpoint
POST https://api.openparser.dev/parse (multipart: request JSON + file)
Upstream model
OpenParser ocr_model "azure-di-layout" (Azure prebuilt-layout, API 2024-11-30)
Auth
Bearer OpenParser API key, server side
Long jobs
HTTP 202, then poll the job until it succeeds

Options

What OCRRouter sets on the upstream call. You choose the model, format and pages in your request; vendor-specific options are not passed through.

ocr_model
"azure-di-layout" (in the JSON request part)
output_format
"openparser@1"
Idempotency-Key
A fresh UUID per request, required by OpenParser
Client options
None. Clients cannot pass vendor options such as page ranges, features or output format.

Raw response fields

Every field the upstream documents, with its type and meaning taken from the vendor's own reference. Fields marked "read by OCRRouter" are the ones our connector uses; the rest are listed so you know what the engine produces.

What OCRRouter receives: OpenParser openparser@1

OCRRouter never calls Azure directly. OpenParser runs the vendor call and converts the vendor response to its own openparser@1 document, which is the only shape OCRRouter sees. Fields marked used are the ones OCRRouter reads.

Source: OpenParser openparser@1 schema, retrieved 2026-11-19

output_format"openparser@1"
Schema version identifier.Example: "openparser@1"
document_idstring
Stable identifier for the parsed document.
provenanceobject
Provider and model metadata: provider, model, version, operation.Example: {"provider":"microsoft","model":"prebuilt-layout","version":"2024-11-30"}
textstring, read by OCRRouter
Plain text of the whole document in reading order. Spans index into it.
markdownstring, read by OCRRouter
Best-effort Markdown rendering of the whole document. Used as the fallback when elements carry no text.Example: "# Invoice 1042\n\nItem Qty"
pages[]DocumentPage[], read by OCRRouter
Coordinate space and reading order of each page.
pages[].numbernumber, read by OCRRouter
One-based page number. Numbers are contiguous.Example: 1
pages[].widthnumber, read by OCRRouter
Page width in the page's coordinate unit.
pages[].heightnumber, read by OCRRouter
Page height in the page's coordinate unit.
pages[].unit"pixel" | "point" | "inch" | "normalized", read by OCRRouter
Coordinate unit for every bbox on the page.Example: "pixel"
pages[].element_idsstring[], read by OCRRouter
IDs of all elements on the page, including lines and words.
pages[].reading_orderstring[], read by OCRRouter
Ordered subset of element_ids, limited to top-level content.
pages[].confidenceConfidence
Optional page-level OCR aggregate.
pages[].qualityPageQuality
Optional image-quality scores, defects and metrics, separate from confidence.
elements[]DocumentElement[], read by OCRRouter
Text, tables, figures and other semantic content.
elements[].idstring, read by OCRRouter
Element ID, referenced from pages[].reading_order.
elements[].kindstring, read by OCRRouter
One of text, table, figure, formula, key_value, query_answer, section, selection_mark, signature, barcode, link, stamp, other.Example: "text"
elements[].rolestring, read by OCRRouter
For text elements: heading, paragraph, line, word, page_header, page_footer and similar.
elements[].textstring, read by OCRRouter
Element text.
elements[].markdownstring, read by OCRRouter
Element text as Markdown, when provided. Read before text.
elements[].spans[]{start,end}[]
UTF-16 offsets into the document text, end exclusive.
elements[].locations[].page_numbernumber, read by OCRRouter
Page the location is on.Example: 1
elements[].locations[].bbox{left,top,right,bottom}, read by OCRRouter
Axis-aligned box in page-local coordinates; right and bottom are exclusive.
elements[].locations[].polygon{x,y}[]
Tighter outline when the source provides one.
elements[].confidence.scorenumber, read by OCRRouter
Score normalised to 0 to 1.
elements[].confidence.scopestring
What the score measures: detection, recognition, classification, geometry, answer or quality.
elements[].confidence.calibratedboolean
Defaults to false. Uncalibrated scores should not be compared across providers or scopes.
elements[].confidence.source_valuenumber
The provider's original score.
elements[].confidence.source_scalestring
Scale of the original score: zero_to_one, zero_to_hundred, log_probability or unknown.
elements[].row_countnumber, read by OCRRouter
Table rows.
elements[].column_countnumber, read by OCRRouter
Table columns.
elements[].cells[].row_indexnumber, read by OCRRouter
Cell row position.
elements[].cells[].column_indexnumber, read by OCRRouter
Cell column position.
elements[].cells[].row_spannumber
Rows the cell covers.
elements[].cells[].column_spannumber
Columns the cell covers.
elements[].cells[].textstring, read by OCRRouter
Cell text.
relations[]DocumentRelation[]
Directed links between elements, such as containment and captions.Example: {"type":"contains","from_id":"...","to_id":"..."}
text_annotations[]TextAnnotation[]
Language and style metadata over spans of text.
assets[]DocumentAsset[]
Page and figure images.

Real response captured 2026-11-19 from POST /parse on a small 600 x 260 PNG. Trimmed to the first two reading-order elements, one relation, and the first element IDs.

{"output_format": "openparser@1","document_id": "b1c4a829aacf83ef0491f8b0f0a48e5c","provenance": {"provider": "microsoft","model": "prebuilt-layout","version": "2024-11-30","operation": "analyze_document"},"text": "# Invoice 1042\n\nItem Qty\n\nPaper\n\n2\n","markdown": "# Invoice 1042\n\nItem Qty\n\nPaper\n\n2","pages": [{"number": 1,"source_page_number": 1,"width": 600,"height": 260,"unit": "pixel","rotation_degrees": 0.9771592020988464,"languages": [],"element_ids": ["azure-paragraph-0","azure-paragraph-1","azure-paragraph-2","azure-paragraph-3","..."],"reading_order": ["azure-paragraph-0","azure-paragraph-1","azure-paragraph-2","azure-paragraph-3"]}],"elements": [{"id": "azure-paragraph-0","locations": [{"page_number": 1,"bbox": { "left": 29, "top": 20, "right": 192, "bottom": 45 },"polygon": [{ "x": 29, "y": 21 },{ "x": 192, "y": 20 },{ "x": 192, "y": 44 },{ "x": 30, "y": 45 }]}],"source": {"native_id": "/paragraphs/0","native_type": "paragraph","native_label": "title"},"kind": "text","role": "document_title","text": "Invoice 1042","spans": [{ "start": 0, "end": 14 }],"languages": []},{"id": "azure-paragraph-1","locations": [{"page_number": 1,"bbox": { "left": 30, "top": 69, "right": 178, "bottom": 98 },"polygon": [{ "x": 30, "y": 69 },{ "x": 178, "y": 71 },{ "x": 177, "y": 98 },{ "x": 30, "y": 95 }]}],"source": { "native_id": "/paragraphs/1", "native_type": "paragraph" },"kind": "text","role": "paragraph","text": "Item Qty","spans": [{ "start": 16, "end": 24 }],"languages": []}],"text_annotations": [],"relations": [{"type": "contains","from_id": "azure-page-1-line-0","to_id": "azure-page-1-word-0"}],"assets": []}

The vendor's own response: Azure AnalyzeResult

What Azure returns to OpenParser from prebuilt-layout. OCRRouter never sees this JSON; it is shown so you know where the openparser@1 values come from (paragraphs become text elements, paragraph role title becomes document_title).

Source: Azure Get Analyze Result REST reference, retrieved 2026-11-19

statusstring
Operation status: notStarted, running, succeeded or failed.Example: "succeeded"
createdDateTimedate-time
UTC time the analyze operation was submitted.
lastUpdatedDateTimedate-time
UTC time the status was last updated.
errorDocumentIntelligenceError
Error encountered during analysis.
analyzeResult.apiVersionstring
API version used to produce the result.Example: "2024-11-30"
analyzeResult.modelIdstring
Document model ID used to produce the result.Example: "prebuilt-layout"
analyzeResult.stringIndexTypestring
Method used to compute string offset and length.Example: "textElements"
analyzeResult.contentstring
Concatenation of all textual and visual elements in reading order.
analyzeResult.contentFormatstring
Format of content: text or markdown (set with outputContentFormat=markdown).Example: "text"
analyzeResult.pages[].pageNumberinteger
1-based page number.Example: 1
analyzeResult.pages[].anglenumber
Orientation angle, showing whether the page is rotated.Example: 0
analyzeResult.pages[].widthnumber
Page width in unit.Example: 8.5
analyzeResult.pages[].heightnumber
Page height in unit.Example: 11
analyzeResult.pages[].unitstring
Unit of width, height and polygons: inch for PDF, pixel for images.Example: "inch"
analyzeResult.pages[].words[]DocumentWord[]
Words with content, polygon, confidence and span.Example: {"content":"CONTOSO","confidence":1}
analyzeResult.pages[].lines[]DocumentLine[]
Lines with content, polygon and spans.
analyzeResult.pages[].selectionMarks[]DocumentSelectionMark[]
Check boxes with state (selected or unselected), polygon, confidence and span.Example: "unselected"
analyzeResult.pages[].spans[]{offset,length}[]
Where the page sits in content.
analyzeResult.paragraphs[].rolestring
Paragraph role: title, sectionHeading, footnote, pageHeader, pageFooter or pageNumber. Absent for body text.Example: "sectionHeading"
analyzeResult.paragraphs[].contentstring
Paragraph text.
analyzeResult.paragraphs[].boundingRegions[]BoundingRegion[]
pageNumber plus polygon: x, y vertices clockwise from the left, relative to the page's top-left.
analyzeResult.tables[].rowCountinteger
Number of rows.Example: 4
analyzeResult.tables[].columnCountinteger
Number of columns.Example: 8
analyzeResult.tables[].cells[].kindstring
Cell kind, for example columnHeader; content cells have no special kind.Example: "columnHeader"
analyzeResult.tables[].cells[].rowIndexinteger
Row of the cell, 0-based.Example: 0
analyzeResult.tables[].cells[].columnIndexinteger
Column of the cell, 0-based.Example: 0
analyzeResult.tables[].cells[].rowSpaninteger
Rows the cell spans.Example: 1
analyzeResult.tables[].cells[].columnSpaninteger
Columns the cell spans.Example: 1
analyzeResult.tables[].cells[].contentstring
Cell text.Example: "DATE"
analyzeResult.figures[]DocumentFigure[]
Figures with id, boundingRegions, spans, elements, caption and footnotes.
analyzeResult.sections[]DocumentSection[]
Hierarchical sections of the document.
analyzeResult.styles[]DocumentStyle[]
Font styles, including isHandwritten for handwritten spans.Example: {"isHandwritten":true}
analyzeResult.keyValuePairs[]DocumentKeyValuePair[]
Extracted key-value pairs.
analyzeResult.languages[]DocumentLanguage[]
Detected languages.
analyzeResult.documents[]AnalyzedDocument[]
Extracted documents (prebuilt and custom models).
analyzeResult.warnings[]DocumentIntelligenceWarning[]
Warnings encountered.

Abridged from Microsoft's documented sample response for Get Analyze Result (API 2024-11-30). That sample uses prebuilt-invoice; prebuilt-layout returns the same AnalyzeResult shape.

{"status": "succeeded","createdDateTime": "2021-09-24T13:00:46Z","lastUpdatedDateTime": "2021-09-24T13:00:49Z","analyzeResult": {"apiVersion": "2024-11-30","modelId": "prebuilt-invoice","stringIndexType": "textElements","contentFormat": "text","content": "CONTOSO LTD.\nINVOICE\nContoso Headquarters\nINVOI…","pages": [{"pageNumber": 1,"angle": 0,"width": 8.5,"height": 11,"unit": "inch","words": [{"content": "CONTOSO","polygon": [0.5911,0.6857,1.7451,0.6857,1.7451,0.8664,0.5911,0.8664],"confidence": 1,"span": { "offset": 0, "length": 7 }}],"selectionMarks": [{"state": "unselected","polygon": [4.3231,5.6279,4.7562,5.6279,4.7562,5.8654,4.3231,5.8654],"confidence": 0.212,"span": { "offset": 946, "length": 12 }}],"lines": [{"content": "CONTOSO LTD.","polygon": [0.5911,0.6857,2.3181,0.6857,2.3181,0.8664,0.5911,0.8664],"spans": [{ "offset": 0, "length": 12 }]}],"spans": [{ "offset": 0, "length": 958 }]}],"tables": [{"rowCount": 4,"columnCount": 8,"cells": [{"kind": "columnHeader","rowIndex": 0,"columnIndex": 0,"rowSpan": 1,"columnSpan": 1,"content": "DATE","boundingRegions": [{"pageNumber": 1,"polygon": [0.4932, 5.329, 1.3011, 5.3213]}]}]}]}}

How it maps to OCRRouter output

pages[] sorted by number
One Markdown string and one layout page per page
pages[].reading_order (else element_ids)
Block order on the page
elements[].kind + role containing table
table block
kind/role containing formula or equation
formula block
kind/role containing figure, image or picture
image block
kind/role containing heading or title
heading block, text prefixed with # (level 1 to 6)
anything else
paragraph block
elements[].markdown, else text
Block text
cells[] with row_count/column_count
GitHub Markdown table, first row as header
locations[].bbox on that page
bbox [left, top, right, bottom] in pages[].unit
pages[].width/height/unit
Layout page size and unit
confidence.score
Block confidence 0 to 1, or null
No element text on any page
Document markdown (else text) returned as page 1, no layout
HTTP 202 with job
Poll GET /jobs/{id} (or Location) with backoff 1 to 5 s until succeeded or failed
Azure paragraph role title
OpenParser role document_title, OCRRouter heading
Azure paragraphs
OpenParser text elements with role paragraph

The unified output is described in Response format.

Example request

curl https://api.ocrrouter.ai/v1/ocr \
  -H "Authorization: Bearer $OCRROUTER_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "azure/document-intelligence",
    "file_url": "https://example.com/invoice.pdf",
    "format": "json"
  }'

Sources

Vendor documentation read on 2026-11-19.