Merge pull request #2978 from MicrosoftDocs/main

2/14/2025 PM Publish

Author: Taojunshen

.whatsnew.json

@@ -0,0 +1,35 @@
+{ 
+    "$schema": "https://whatsnewapi.azurewebsites.net/schema",
+    "docSetProductName": "Azure AI Services",
+    "rootDirectory": "articles/",
+    "docLinkSettings": {
+        "linkFormat": "relative",
+        "relativeLinkPrefix": "../"
+    },
+    "navigationOptions": {
+        "maximumNumberOfArticles": 5,
+        "tocParentNode": "Latest documentation updates",
+        "repoTocFolder": "articles/ai-services",
+        "indexParentNode": "Latest documentation updates",
+        "repoIndexFolder": "articles/ai-services"
+    },
+    "areas": [
+        {
+            "names": [ "ai-services" ],
+            "heading": "Azure AI Services"
+        },
+        {
+            "names": [ "ai-studio" ],
+            "heading": "Azure AI Foundry"
+        },
+        {
+            "names": [ "machine-learning" ],
+            "heading": "Azure Machine Learning"
+        },
+        {
+            "names": [ "search" ],
+            "heading": "Azure AI Search"
+    
+        }
+    ]
+}

articles/ai-foundry/model-inference/how-to/use-embeddings.md

@@ -48,5 +48,5 @@ zone_pivot_groups: azure-ai-inference-samples
 
 ## Related content
 
-* [Use embeddings models](use-embeddings.md)
+* [Use image embeddings models](use-image-embeddings.md)
 * [Azure AI Model Inference API](.././reference/reference-model-inference-api.md)

articles/ai-foundry/model-inference/includes/code-create-chat-client-entra.md

@@ -26,7 +26,6 @@ client = ChatCompletionsClient(
     endpoint="https://<resource>.services.ai.azure.com/models",
     credential=DefaultAzureCredential(),
     credential_scopes=["https://cognitiveservices.azure.com/.default"],
-    model="mistral-large-2407",
 )
 ```
 
@@ -50,7 +49,6 @@ const clientOptions = { credentials: { "https://cognitiveservices.azure.com" } }
 const client = new ModelClient(
     "https://<resource>.services.ai.azure.com/models", 
     new DefaultAzureCredential(),
-    "mistral-large-2407",
     clientOptions,
 );
 ```
@@ -80,15 +78,14 @@ using Azure.AI.Inference;
 Then, you can use the package to consume the model. The following example shows how to create a client to consume chat completions with Entra ID:
 
 ```csharp
-var credential = new DefaultAzureCredential();
+TokenCredential credential = new DefaultAzureCredential();
 AzureAIInferenceClientOptions clientOptions = new AzureAIInferenceClientOptions();
 BearerTokenAuthenticationPolicy tokenPolicy = new BearerTokenAuthenticationPolicy(credential, new string[] { "https://cognitiveservices.azure.com/.default" });
 clientOptions.AddPolicy(tokenPolicy, HttpPipelinePosition.PerRetry);
 
 ChatCompletionsClient client = new ChatCompletionsClient(
     new Uri("https://<resource>.services.ai.azure.com/models"),
     credential,
-    "mistral-large-2407",
     clientOptions.
 );
 ```
@@ -117,7 +114,6 @@ TokenCredential defaultCredential = new DefaultAzureCredentialBuilder().build();
 ChatCompletionsClient client = new ChatCompletionsClientBuilder()
     .credential(defaultCredential)
     .endpoint("https://<resource>.services.ai.azure.com/models")
-    .model("mistral-large-2407")
     .buildClient();
 ```
 

articles/ai-foundry/model-inference/includes/use-chat-completions/csharp.md

@@ -47,18 +47,23 @@ First, create the client to consume the model. The following code uses an endpoi
 ChatCompletionsClient client = new ChatCompletionsClient(
     new Uri(Environment.GetEnvironmentVariable("AZURE_INFERENCE_ENDPOINT")),
     new AzureKeyCredential(Environment.GetEnvironmentVariable("AZURE_INFERENCE_CREDENTIAL")),
-    "mistral-large-2407"
 );
 ```
 
 If you have configured the resource to with **Microsoft Entra ID** support, you can use the following code snippet to create a client.
 
 
 ```csharp
+TokenCredential credential = new DefaultAzureCredential(includeInteractiveCredentials: true);
+AzureAIInferenceClientOptions clientOptions = new AzureAIInferenceClientOptions();
+BearerTokenAuthenticationPolicy tokenPolicy = new BearerTokenAuthenticationPolicy(credential, new string[] { "https://cognitiveservices.azure.com/.default" });
+
+clientOptions.AddPolicy(tokenPolicy, HttpPipelinePosition.PerRetry);
+
 client = new ChatCompletionsClient(
     new Uri(Environment.GetEnvironmentVariable("AZURE_INFERENCE_ENDPOINT")),
-    new DefaultAzureCredential(includeInteractiveCredentials: true),
-    "mistral-large-2407"
+    credential,
+    clientOptions,
 );
 ```
 
@@ -73,6 +78,7 @@ ChatCompletionsOptions requestOptions = new ChatCompletionsOptions()
         new ChatRequestSystemMessage("You are a helpful assistant."),
         new ChatRequestUserMessage("How many languages are in the world?")
     },
+    Model = "mistral-large-2407",
 };
 
 Response<ChatCompletions> response = client.Complete(requestOptions);
@@ -122,7 +128,8 @@ static async Task StreamMessageAsync(ChatCompletionsClient client)
             new ChatRequestSystemMessage("You are a helpful assistant."),
             new ChatRequestUserMessage("How many languages are in the world? Write an essay about it.")
         },
-        MaxTokens=4096
+        MaxTokens=4096,
+        Model = "mistral-large-2407",
     };
 
     StreamingResponse<StreamingChatCompletionsUpdate> streamResponse = await client.CompleteStreamingAsync(requestOptions);
@@ -168,6 +175,7 @@ requestOptions = new ChatCompletionsOptions()
         new ChatRequestSystemMessage("You are a helpful assistant."),
         new ChatRequestUserMessage("How many languages are in the world?")
     },
+    Model = "mistral-large-2407",
     PresencePenalty = 0.1f,
     FrequencyPenalty = 0.8f,
     MaxTokens = 2048,
@@ -202,7 +210,8 @@ requestOptions = new ChatCompletionsOptions()
             "How many languages are in the world?"
         )
     },
-    ResponseFormat = new ChatCompletionsResponseFormatJSON()
+    ResponseFormat = new ChatCompletionsResponseFormatJsonObject(),
+    Model = "mistral-large-2407",
 };
 
 response = client.Complete(requestOptions);
@@ -221,6 +230,7 @@ requestOptions = new ChatCompletionsOptions()
         new ChatRequestSystemMessage("You are a helpful assistant."),
         new ChatRequestUserMessage("How many languages are in the world?")
     },
+    Model = "mistral-large-2407",
     AdditionalProperties = { { "logprobs", BinaryData.FromString("true") } },
 };
 
@@ -295,7 +305,7 @@ var chatHistory = new List<ChatRequestMessage>(){
         new ChatRequestUserMessage("When is the next flight from Miami to Seattle?")
     };
 
-requestOptions = new ChatCompletionsOptions(chatHistory);
+requestOptions = new ChatCompletionsOptions(chatHistory, model: "mistral-large-2407");
 requestOptions.Tools.Add(getFlightTool);
 requestOptions.ToolChoice = ChatCompletionsToolChoice.Auto;
 
@@ -377,6 +387,7 @@ try
                 "Chopping tomatoes and cutting them into cubes or wedges are great ways to practice your knife skills."
             ),
         },
+        Model = "mistral-large-2407",
     };
 
     response = client.Complete(requestOptions);
@@ -436,6 +447,7 @@ ChatCompletionsOptions requestOptions = new ChatCompletionsOptions()
         ]),
     },
     MaxTokens=2048,
+    Model = "phi-3.5-vision-instruct",
 };
 
 var response = client.Complete(requestOptions);
@@ -456,7 +468,7 @@ Console.WriteLine($"\tCompletion tokens: {response.Value.Usage.CompletionTokens}
 
 ```console
 ASSISTANT: The chart illustrates that larger models tend to perform better in quality, as indicated by their size in billions of parameters. However, there are exceptions to this trend, such as Phi-3-medium and Phi-3-small, which outperform smaller models in quality. This suggests that while larger models generally have an advantage, there might be other factors at play that influence a model's performance.
-Model: mistral-large-2407
+Model: phi-3.5-vision-instruct
 Usage: 
   Prompt tokens: 2380
   Completion tokens: 126

articles/ai-foundry/model-inference/includes/use-chat-reasoning/csharp.md

@@ -42,8 +42,7 @@ First, create the client to consume the model. The following code uses an endpoi
 ```csharp
 ChatCompletionsClient client = new ChatCompletionsClient(
     new Uri("https://<resource>.services.ai.azure.com/models"),
-    new AzureKeyCredential(Environment.GetEnvironmentVariable("AZURE_INFERENCE_CREDENTIAL")),
-    "DeepSeek-R1"
+    new AzureKeyCredential(Environment.GetEnvironmentVariable("AZURE_INFERENCE_CREDENTIAL"))
 );
 ```
 
@@ -53,10 +52,16 @@ ChatCompletionsClient client = new ChatCompletionsClient(
 If you have configured the resource to with **Microsoft Entra ID** support, you can use the following code snippet to create a client.
 
 ```csharp
+TokenCredential credential = new DefaultAzureCredential(includeInteractiveCredentials: true);
+AzureAIInferenceClientOptions clientOptions = new AzureAIInferenceClientOptions();
+BearerTokenAuthenticationPolicy tokenPolicy = new BearerTokenAuthenticationPolicy(credential, new string[] { "https://cognitiveservices.azure.com/.default" });
+
+clientOptions.AddPolicy(tokenPolicy, HttpPipelinePosition.PerRetry);
+
 client = new ChatCompletionsClient(
     new Uri("https://<resource>.services.ai.azure.com/models"),
-    new DefaultAzureCredential(includeInteractiveCredentials: true),
-    "DeepSeek-R1"
+    credential,
+    clientOptions,
 );
 ```
 
@@ -70,6 +75,7 @@ ChatCompletionsOptions requestOptions = new ChatCompletionsOptions()
     Messages = {
         new ChatRequestUserMessage("How many languages are in the world?")
     },
+    Model = "deepseek-r1",
 };
 
 Response<ChatCompletions> response = client.Complete(requestOptions);
@@ -148,7 +154,8 @@ static async Task StreamMessageAsync(ChatCompletionsClient client)
         Messages = {
             new ChatRequestUserMessage("How many languages are in the world?")
         },
-        MaxTokens=4096
+        MaxTokens=4096,
+        Model = "deepseek-r1",
     };
 
     StreamingResponse<StreamingChatCompletionsUpdate> streamResponse = await client.CompleteStreamingAsync(requestOptions);
@@ -225,6 +232,7 @@ try
                 "Chopping tomatoes and cutting them into cubes or wedges are great ways to practice your knife skills."
             ),
         },
+        Model = "deepseek-r1",
     };
 
     response = client.Complete(requestOptions);

articles/ai-foundry/model-inference/includes/use-embeddings/csharp.md

@@ -49,19 +49,23 @@ First, create the client to consume the model. The following code uses an endpoi
 ```csharp
 EmbeddingsClient client = new EmbeddingsClient(
     new Uri(Environment.GetEnvironmentVariable("AZURE_INFERENCE_ENDPOINT")),
-    new AzureKeyCredential(Environment.GetEnvironmentVariable("AZURE_INFERENCE_CREDENTIAL")),
-    "text-embedding-3-small"
+    new AzureKeyCredential(Environment.GetEnvironmentVariable("AZURE_INFERENCE_CREDENTIAL"))
 );
 ```
 
-If you have configured the resource to with **Microsoft Entra ID** support, you can use the following code snippet to create a client.
-
+If you configured the resource to with **Microsoft Entra ID** support, you can use the following code snippet to create a client. Note that here `includeInteractiveCredentials` is set to `true` only for demonstration purposes so authentication can happen using the web browser. On production workloads, you should remove such parameter.
 
 ```csharp
+TokenCredential credential = new DefaultAzureCredential(includeInteractiveCredentials: true);
+AzureAIInferenceClientOptions clientOptions = new AzureAIInferenceClientOptions();
+BearerTokenAuthenticationPolicy tokenPolicy = new BearerTokenAuthenticationPolicy(credential, new string[] { "https://cognitiveservices.azure.com/.default" });
+
+clientOptions.AddPolicy(tokenPolicy, HttpPipelinePosition.PerRetry);
+
 client = new EmbeddingsClient(
-    new Uri(Environment.GetEnvironmentVariable("AZURE_INFERENCE_ENDPOINT")),
-    new DefaultAzureCredential(includeInteractiveCredentials: true),
-    "text-embedding-3-small"
+    new Uri("https://<resource>.services.ai.azure.com/models"),
+    credential,
+    clientOptions,
 );
 ```
 
@@ -75,6 +79,7 @@ EmbeddingsOptions requestOptions = new EmbeddingsOptions()
     Input = {
         "The ultimate answer to the question of life"
     },
+    Model = "text-embedding-3-small"
 };
 
 Response<EmbeddingsResult> response = client.Embed(requestOptions);
@@ -104,6 +109,7 @@ EmbeddingsOptions requestOptions = new EmbeddingsOptions()
         "The ultimate answer to the question of life", 
         "The largest planet in our solar system is Jupiter"
     },
+    Model = "text-embedding-3-small"
 };
 
 Response<EmbeddingsResult> response = client.Embed(requestOptions);
@@ -129,7 +135,12 @@ The following example shows how to create embeddings that are used to create an
 var input = new List<string> { 
     "The answer to the ultimate question of life, the universe, and everything is 42"
 };
-var requestOptions = new EmbeddingsOptions(input, EmbeddingInputType.DOCUMENT);
+var requestOptions = new EmbeddingsOptions()
+{
+    Input = input,
+    InputType = EmbeddingInputType.DOCUMENT, 
+    Model = "text-embedding-3-small"
+};
 
 Response<EmbeddingsResult> response = client.Embed(requestOptions);
 ```
@@ -141,7 +152,12 @@ When you work on a query to retrieve such a document, you can use the following
 var input = new List<string> { 
     "What's the ultimate meaning of life?"
 };
-var requestOptions = new EmbeddingsOptions(input, EmbeddingInputType.QUERY);
+var requestOptions = new EmbeddingsOptions()
+{
+    Input = input,
+    InputType = EmbeddingInputType.QUERY,
+    Model = "text-embedding-3-small"
+};
 
 Response<EmbeddingsResult> response = client.Embed(requestOptions);
 ```