Azure Open AI Resource URi to configure an Integrated vectorizer in a search Index of Azure AI Search

[danielperezefremova-tomtom]

Hi!

Background

I am working on integrating Open AI Embeddings computation with Azure AI Search Index.

I am following the tutorial from Microsoft site: Configure a vectorizer in a search index

Then, I realized that the URi is slightly different compared to the endpoint we use to query chatgpt:

• REQUIRED: https://XXXXXX**.openai.azure.com**/
• CURRENT: https://api.chatgpt.tomtom-global.com/

When I try to declare the index

{
  "name": "osm_map_features",
  "defaultScoringProfile": null,
  "fields": [
    {
      "name": "id",
      "type": "Edm.String",
      "key": true
    },
    {
      "name": "key",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "sortable": true,
      "facetable": true
    },
    {
      "name": "value",
      "type": "Edm.String",
      "searchable": true,
      "filterable": true,
      "retrievable": true,
      "sortable": true,
      "facetable": true
    },
    {
      "name": "description",
      "type": "Edm.String",
      "searchable": true,
      "filterable": false,
      "retrievable": true,
      "sortable": false,
      "facetable": false
    },
    {
      "name": "descriptionVector",
      "type": "Collection(Edm.Single)",
      "searchable": true,
      "filterable": false,
      "retrievable": true,
      "sortable": false,
      "facetable": false,
      "key": false,
      "indexAnalyzer": null,
      "searchAnalyzer": null,
      "analyzer": null,
      "normalizer": null,
      "dimensions": 1536,
      "vectorSearchProfile": "openai-vectorSearch",
      "synonymMaps": []
    }
    ],
  "vectorSearch": {
     "profiles": [
        {
          "name": "openai-vectorSearch",
          "algorithm": "vectorSearch-algorithm",
          "vectorizer": "openai-embeddings"
        }
     ],
     "algorithms": [
       {
          "name": "vectorSearch-algorithm",
          "kind": "hnsw",
          "hnswParameters": {
            "m": 4,
            "metric": "cosine",
            "efConstruction": 400,
            "efSearch": 500
           }
       }
      ],
      "vectorizers": [
        {
          "name": "openai-embeddings",
          "kind": "azureOpenAI",
          "azureOpenAIParameters": {
            "resourceUri": "https://api.chatgpt.tomtom-global.com/",
            "deploymentId": "dep-embed-ada",
            "apiKey": "XXXX"
          }
        }
      ]
   }
}

The Azure AI Search service throws:

Error: The request is invalid. Details: definition : Error in Vectorizer 'openai-embeddings' : Invalid resourceUri. Provide a valid resourceUri for the Azure Open AI service.

QUESTION

I suppose the URi is not correct. As Chatty is using this service, do you know the appropiate URi to set up the AI Search service?