What value do we provide in authidentity field in vectorizers while creating index or skillset?

Question

What value do we provide in authidentity field in vectorizers while creating index or skillset?

imgkj 45

I want create a index and skillset but while creating the index with JSON, I am not sure what value do I need to provide in vestorizers for authidentiy. I don't want to use api keys and am looking to use managed identity to serve the purpose. How do I proceed for this use case?

Can I use only authidentity and leave api keys blank in the JSON to achieve this? if yes, what value I can provie in authidentity field? Please share any example to understand this if possible.

Here's the snippet of the reference JSON from MS learn portal. User's image

I want to know how to provide this authidentity field value and will this work if I remove the api key or leave it blank. I just want to use managed identity for this.

1 answer

Your answer

Answer 1

In Azure AI Search, to use managed identity instead of an API key in vectorizers, you can indeed use the authIdentity field, while omitting the apiKey. To do this, specify authIdentity with the value ManagedIdentity. :


{

    "profiles": [

        {

            "name": "vector-nasa-ebook-text-profile",

            "algorithm": "vector-nasa-ebook-text-algorithm",

            "vectorizer": "vector-nasa-ebook-text-vectorizer"

        }

    ],

    "vectorizers": [

        {

            "name": "vector-nasa-ebook-text-vectorizer",

            "kind": "azureOpenAI",

            "azureOpenAIParameters": {

                "resourceUri": "https://my-fake-azure-openai-resource.openai.azure.com/",

                "deploymentId": "text-embedding-ada-002",

                "modelName": "text-embedding-ada-002",

                "authIdentity": "ManagedIdentity"

            },

            "customWebApiParameters": null

        }

    ]

}

But before verify that the managed identity you are using has the necessary permissions on the Azure OpenAI resource. Usually, this requires at least the "Cognitive Services User" role on the target resource.

The apiKey can be omitted or left as null when authIdentity is set to ManagedIdentity.

For further guidance, you can refer to the Azure Cognitive Search documentation or the Azure OpenAI integration guides.

imgkj 45 Reputation points

2024-10-28T10:06:24.23+00:00

Thanks for the info above.

Adding to this I have another query - while creating a skillset, I see I need to provide Cognitive services details in JSON(attached the screenshot from a MS doc). Is there another way of doing this except providing the key field in the cognitive service block? Is there an option to not provide the key in JSON. can we omit cognitive service block? Also, what are billable and non billable skills sets and what if I don't want to use the billable ones.
imgkj 45 Reputation points

2024-10-28T10:09:16.8666667+00:00

Thanks for the help above.

Adding to that I have another query - While creating skill set, I see a block Cognitive service which has an attribute 'key'. Is there another way of authenticating than providing key as a part of the JSON input? I also need to understand what billable and non billable skillsets are. What if I want to use non billable skill set? will I still need to provide the cognitive service block?

Share via

What value do we provide in authidentity field in vectorizers while creating index or skillset?

1 answer

Your answer