ModelNotFound error

Question

ModelNotFound error

Flavius Sfirlea 0

Hello,

I have created a custom extraction model, labeled the fields, trained it and tested it, however when I am calling the API I am getting the following error:

HTTP/1.1 404 NotFound

ms-azure-ai-errorcode: NotFound

x-ms-error-code: NotFound

x-envoy-upstream-service-time: 32

apim-request-id: 1e61a06e-1a98-48a4-ad45-f6e197848c09

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

x-content-type-options: nosniff

x-ms-region: West Europe

Content-Length: 146

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

Date: Wed, 10 Dec 2025 12:08:19 GMT

{

"error": {

"code": "NotFound",

"message": "Resource not found.",

"innererror": {

  "code": "ModelNotFound",

  "message": "The requested model was not found."

}

}

If I am calling the API using a prebuilt model seems to work and I am getting a 200 response.

However, when I'm using the custom extraction model that I have created I am getting the error from above.

1 answer

Your answer

Answer 1

Hi Flavius Sfirlea,

Thank you for reaching out on the Microsoft Q&A.
It looks like you're running into a "ModelNotFound" error when trying to call your custom extraction model via the API. This can be frustrating, but here are some steps you can take to troubleshoot the issue:

Verify Model and Endpoint Configuration:
- Double-check that you're using the correct model ID in your API call. It's easy to have a typo that can cause this error.
- Make sure the endpoint URL you’re using matches the region where your Document Intelligence resource is deployed.
Check Azure Resource Availability:
- Go to the Azure status page to see if there are any outages or issues affecting the Azure Cognitive Services in your region.
Review API Permissions:
- Confirm that your application has the right permissions to access the Document Intelligence API. Check the Access control (IAM) settings in the Azure portal.
Check for Deployment Status:
- Ensure that the custom model is deployed and active in the Azure portal. Sometimes, models can fail to deploy properly.
Authentication Details:
- Make sure your authentication credentials (like tenant ID, client ID, and secret) are configured correctly, and test the connection to confirm they’re working.
Experiment with Prebuilt Models:
Since prebuilt models work for you, it’s good to know that your API is functioning correctly in that scenario. This points more towards the custom model configuration rather than a systemic issue.

Relevant Documentation:

I Hope this helps. Do let me know if you have any further queries.

If this answers your query, please do click Accept Answer and Yes for was this answer helpful.

Thank you!

Share via

ModelNotFound error

1 answer

Your answer