Is there a way to have the markdown property omitted from the response of the Get analyze result endpoint for my analyzer?

Question

Is there a way to have the markdown property omitted from the response of the Get analyze result endpoint for my analyzer?

Miles 0

{
	"id": "ea138e29-41ba-444c-ac0a-524e63fb95fd",
	"status": "Succeeded",
	"result": {
		"analyzerId": "my_analyzer",
		"apiVersion": "2025-11-01",
		"createdAt": "2026-08-13T01:54:02Z",
		"stringEncoding": "codePoint",
		"warnings": [],
		"contents": [
			{
				"path": "input1",
				"markdown": "Some Date: 2025-07-02\n\nMore Data: Hello, world\n",
				"fields": {
					"SomeDate": {
						"type": "date",
						"valueDate": "2025-07-02",
						"spans": [
							{
								"offset": 1175,
								"length": 10
							}
						],
						"confidence": 0.952,
						"source": "D(1,961.0000,715.0000,1075.0000,717.0000,1075.0000,738.0000,961.0000,737.0000)"
					}
				},
				"kind": "document",
				"startPageNumber": 1,
				"endPageNumber": 1,
				"unit": "pixel",
				"pages": [
					{
						"pageNumber": 1,
						"angle": 0.1182302,
						"width": 1387,
						"height": 1934
					}
				],
				"analyzerId": "my_analyzer",
				"mimeType": "image/jpeg"
			}
		]
	},
	"usage": {
		"documentPagesStandard": 1,
		"contextualizationTokens": 1000,
		"tokens": {
			"gpt-4.1-input": 9793,
			"gpt-4.1-output": 461
		}
	}
}

Miles 0

I can recreate the the analyzer if there is a configuration that can be included.

PUT "{endpoint}/contentunderstanding/analyzers/{analyzerId}?api-version=2025-11-01"

{
  "description": "",
  "baseAnalyzerId": "prebuilt-document",
  "config": {
    "returnDetails": false,
    "enableOcr": true,
    "enableLayout": true,
    "enableFormula": true,
    "enableFigureDescription": false,
    "enableFigureAnalysis": false,
    "chartFormat": "chartjs",
    "tableFormat": "html",
    "enableSegment": false,
    "omitContent": false,
    "segmentPerPage": false,
    "annotationFormat": "markdown",
    "includeMarkdown": false //something like this
  },
  "models": {
    "completion": "gpt-4.1",
    "embedding": "text-embedding-3-large"
  },
  "fieldSchema": {
    "fields": {
      "SomeDate": {
        "type": "date",
        "method": "extract",
        "description": "The date field"
      }
    }
  }
}

1 answer

Your answer

Miles 0 Reputation points

2026-08-14T02:30:57.1433333+00:00

I can recreate the the analyzer if there is a configuration that can be included.

PUT "{endpoint}/contentunderstanding/analyzers/{analyzerId}?api-version=2025-11-01"

{ "description": "", "baseAnalyzerId": "prebuilt-document", "config": { "returnDetails": false, "enableOcr": true, "enableLayout": true, "enableFormula": true, "enableFigureDescription": false, "enableFigureAnalysis": false, "chartFormat": "chartjs", "tableFormat": "html", "enableSegment": false, "omitContent": false, "segmentPerPage": false, "annotationFormat": "markdown", "includeMarkdown": false //something like this }, "models": { "completion": "gpt-4.1", "embedding": "text-embedding-3-large" }, "fieldSchema": { "fields": { "SomeDate": { "type": "date", "method": "extract", "description": "The date field" } } } }

Answer 1

No. The documented Get analyze result response schema includes contents[].markdown as an optional field, and the available request query parameters shown for analyze operations are stringEncoding, processingLocation, and range. No parameter is documented to suppress or omit markdown from the analyzer result response.

If markdown must not be returned to downstream code, the practical approach is to ignore or remove contents[].markdown after receiving the result.

In the .NET SDK preview, LlmInputOptions.IncludeMarkdown exists and defaults to true, but that applies to LLM input options, not the Get analyze result endpoint response schema shown for analyzer results.

References: