Getting 404 error when trying to access custom classification model in new Document Intelligence Studio

Jac Phillipps 0 Reputation points
2025-10-23T14:46:23.15+00:00

I've created a custom classification model in the new Azure Document Intelligence Studio and I have trained it with my relevant documents. I am now trying to access this trained model in my c# service but I am getting a 404 resource not found when using my custom model with the following code

var client = new HttpClient();

client.DefaultRequestHeaders.Add("Ocp-Apim-Subscription-Key", _docIntelligenceConfig.ApiKey);

var url = $"{_docIntelligenceConfig.Endpoint}documentintelligence/documentModels/{_docIntelligenceConfig.ModelId}:analyze?api-version=2024-11-30";

var fileBytes = await File.ReadAllBytesAsync(filePath);

var content = new ByteArrayContent(fileBytes);

content.Headers.ContentType = new System.Net.Http.Headers.MediaTypeHeaderValue("application/pdf");

var response = await client.PostAsync(url, content);

With the endpoint coming from my connected azure portal resource

User's image

User's image

And I've double checked my model id by running an analysis on a document with the trained model and going to the results tab once the analysis has completed

User's image

In addition to this I have double checked my endpoint works as expected as I have updated the code to an example model, which succeeds :

var modelId = "prebuilt-document"; // test with a prebuilt model

var client = new HttpClient();

client.DefaultRequestHeaders.Add("Ocp-Apim-Subscription-Key", _docIntelligenceConfig.ApiKey);

// Use formrecognizer route

var url = $"{_docIntelligenceConfig.Endpoint}formrecognizer/documentModels/{modelId}:analyze?api-version=2023-07-31";

var fileBytes = await File.ReadAllBytesAsync(filePath);

var content = new ByteArrayContent(fileBytes);

content.Headers.ContentType = new System.Net.Http.Headers.MediaTypeHeaderValue("application/pdf");

var response = await client.PostAsync(url, content);

I have also tried simply plugging in my model in place of the "prebuilt-document" model and ensuring that I retrain the model in the same api version as the example but this still comes back with the 404 error.

I am unsure what to try next, any recommendations?

Full Error :

{StatusCode: 404, ReasonPhrase: 'Not Found', Version: 1.1, Content: System.Net.Http.HttpConnectionResponseContent, Headers:

{

ms-azure-ai-errorcode: NotFound

x-ms-error-code: NotFound

x-envoy-upstream-service-time: 23

apim-request-id: id here

Strict-Transport-Security: max-age=31536000; includeSubDomains; preload

X-Content-Type-Options: nosniff

x-ms-region: UK South

Date: Fri, 17 Oct 2025 12:22:50 GMT

Content-Length: 146

Content-Type: application/json; charset=utf-8

}}

Azure AI Document Intelligence
{count} votes

Your answer

Answers can be marked as 'Accepted' by the question author and 'Recommended' by moderators, which helps users know the answer solved the author's problem.