Merge branch 'main' into feature-connectors-anthropic

dotnet/Directory.Packages.props

@@ -6,10 +6,10 @@
   </PropertyGroup>
   <ItemGroup>
     <PackageVersion Include="Azure.AI.Inference" Version="1.0.0-beta.1" />
-    <PackageVersion Include="OpenAI" Version="2.0.0-beta.10" />
+    <PackageVersion Include="OpenAI" Version="2.0.0-beta.11" />
     <PackageVersion Include="System.ClientModel" Version="1.1.0-beta.7" />
     <PackageVersion Include="Azure.AI.ContentSafety" Version="1.0.0" />
-    <PackageVersion Include="Azure.AI.OpenAI" Version="2.0.0-beta.4" />
+    <PackageVersion Include="Azure.AI.OpenAI" Version="2.0.0-beta.5" />
     <PackageVersion Include="Azure.Identity" Version="1.12.0" />
     <PackageVersion Include="Azure.Monitor.OpenTelemetry.Exporter" Version="1.3.0" />
     <PackageVersion Include="Azure.Search.Documents" Version="11.6.0" />

dotnet/docs/OPENAI-CONNECTOR-MIGRATION.md

@@ -53,35 +53,55 @@ tags: `ResultsPerPrompt`,`results_per_prompt`
 
 The `OpenAIFileService` was deprecated in the latest version of the OpenAI Connector. We strongly recommend to update your code to use the new `OpenAIClient.GetFileClient()` for file management operations.
 
-## 5. SemanticKernel MetaPackage
+## 5. OpenAI ChatCompletion custom endpoint
+
+The `OpenAIChatCompletionService` **experimental** constructor for custom endpoints will not attempt to auto-correct the endpoint and use it as is.
+
+We have the two only specific cases where we attempted to auto-correct the endpoint.
+
+1. If you provided `chat/completions` path before. Now those need to be removed as they are added automatically to the end of your original endpoint by `OpenAI SDK`.
+
+   ```diff
+   - http://any-host-and-port/v1/chat/completions
+   + http://any-host-and-port/v1
+   ```
+
+2. If you provided a custom endpoint without any path. We won't be adding the `v1/` as the first path. Now the `v1` path needs to provided as part of your endpoint.
+
+   ```diff
+   - http://any-host-and-port/
+   + http://any-host-and-port/v1
+   ```
+
+## 6. SemanticKernel MetaPackage
 
 To be retro compatible with the new OpenAI and AzureOpenAI Connectors, our `Microsoft.SemanticKernel` meta package changed its dependency to use the new `Microsoft.SemanticKernel.Connectors.AzureOpenAI` package that depends on the `Microsoft.SemanticKernel.Connectors.OpenAI` package. This way if you are using the metapackage, no change is needed to get access to `Azure` related types.
 
-## 6. Contents
+## 7. Contents
 
-### 6.1 OpenAIChatMessageContent
+### 7.1 OpenAIChatMessageContent
 
 - The `Tools` property type has changed from `IReadOnlyList<ChatCompletionsToolCall>` to `IReadOnlyList<ChatToolCall>`.
 
 - Inner content type has changed from `ChatCompletionsFunctionToolCall` to `ChatToolCall`.
 
 - Metadata type `FunctionToolCalls` has changed from `IEnumerable<ChatCompletionsFunctionToolCall>` to `IEnumerable<ChatToolCall>`.
 
-### 6.2 OpenAIStreamingChatMessageContent
+### 7.2 OpenAIStreamingChatMessageContent
 
 - The `FinishReason` property type has changed from `CompletionsFinishReason` to `FinishReason`.
 - The `ToolCallUpdate` property has been renamed to `ToolCallUpdates` and its type has changed from `StreamingToolCallUpdate?` to `IReadOnlyList<StreamingToolCallUpdate>?`.
 - The `AuthorName` property is not initialized because it's not provided by the underlying library anymore.
 
-## 6.3 Metrics for AzureOpenAI Connector
+## 7.3 Metrics for AzureOpenAI Connector
 
 The meter `s_meter = new("Microsoft.SemanticKernel.Connectors.OpenAI");` and the relevant counters still have old names that contain "openai" in them, such as:
 
 - `semantic_kernel.connectors.openai.tokens.prompt`
 - `semantic_kernel.connectors.openai.tokens.completion`
 - `semantic_kernel.connectors.openai.tokens.total`
 
-## 7. Using Azure with your data (Data Sources)
+## 8. Using Azure with your data (Data Sources)
 
 With the new `AzureOpenAIClient`, you can now specify your datasource thru the options and that requires a small change in your code to the new type.
 
@@ -116,41 +136,41 @@ var promptExecutionSettings = new AzureOpenAIPromptExecutionSettings
 };
 ```
 
-## 8. Breaking glass scenarios
+## 9. Breaking glass scenarios
 
 Breaking glass scenarios are scenarios where you may need to update your code to use the new OpenAI Connector. Below are some of the breaking changes that you may need to be aware of.
 
-#### 8.1 KernelContent Metadata
+#### 9.1 KernelContent Metadata
 
 Some of the keys in the content metadata dictionary have changed, you will need to update your code to when using the previous key names.
 
 - `Created` -> `CreatedAt`
 
-#### 8.2 Prompt Filter Results
+#### 9.2 Prompt Filter Results
 
 The `PromptFilterResults` metadata type has changed from `IReadOnlyList<ContentFilterResultsForPrompt>` to `ContentFilterResultForPrompt`.
 
-#### 8.3 Content Filter Results
+#### 9.3 Content Filter Results
 
 The `ContentFilterResultsForPrompt` type has changed from `ContentFilterResultsForChoice` to `ContentFilterResultForResponse`.
 
-#### 8.4 Finish Reason
+#### 9.4 Finish Reason
 
 The FinishReason metadata string value has changed from `stop` to `Stop`
 
-#### 8.5 Tool Calls
+#### 9.5 Tool Calls
 
 The ToolCalls metadata string value has changed from `tool_calls` to `ToolCalls`
 
-#### 8.6 LogProbs / Log Probability Info
+#### 9.6 LogProbs / Log Probability Info
 
 The `LogProbabilityInfo` type has changed from `ChatChoiceLogProbabilityInfo` to `IReadOnlyList<ChatTokenLogProbabilityInfo>`.
 
-#### 8.7 Finish Details, Index, and Enhancements
+#### 9.7 Finish Details, Index, and Enhancements
 
 All of above have been removed.
 
-#### 8.8 Token Usage
+#### 9.8 Token Usage
 
 The Token usage naming convention from `OpenAI` changed from `Completion`, `Prompt` tokens to `Output` and `Input` respectively. You will need to update your code to use the new naming.
 
@@ -172,13 +192,13 @@ The type also changed from `CompletionsUsage` to `ChatTokenUsage`.
 totalTokens: usage?.TotalTokens ?? 0;
 ```
 
-#### 8.9 OpenAIClient
+#### 9.9 OpenAIClient
 
 The `OpenAIClient` type previously was a Azure specific namespace type but now it is an `OpenAI` SDK namespace type, you will need to update your code to use the new `OpenAIClient` type.
 
 When using Azure, you will need to update your code to use the new `AzureOpenAIClient` type.
 
-#### 8.10 Pipeline Configuration
+#### 9.10 Pipeline Configuration
 
 The new `OpenAI` SDK uses a different pipeline configuration, and has a dependency on `System.ClientModel` package. You will need to update your code to use the new `HttpClientPipelineTransport` transport configuration where before you were using `HttpClientTransport` from `Azure.Core.Pipeline`.
 

dotnet/src/Agents/Core/History/ChatHistorySummarizationReducer.cs

@@ -29,7 +29,7 @@ public class ChatHistorySummarizationReducer : IChatHistoryReducer
     /// </summary>
     public const string DefaultSummarizationPrompt =
         """
-        Provide a concise and complete summarizion of the entire dialog that does not exceed 5 sentences
+        Provide a concise and complete summarization of the entire dialog that does not exceed 5 sentences
 
         This summary must always:
         - Consider both user and assistant interactions

dotnet/src/Connectors/Connectors.AzureOpenAI/Core/AzureClientCore.ChatCompletion.cs

@@ -43,7 +43,9 @@ protected override ChatCompletionOptions CreateChatCompletionOptions(
             TopP = (float?)executionSettings.TopP,
             FrequencyPenalty = (float?)executionSettings.FrequencyPenalty,
             PresencePenalty = (float?)executionSettings.PresencePenalty,
+#pragma warning disable OPENAI001 // Type is for evaluation purposes only and is subject to change or removal in future updates. Suppress this diagnostic to proceed.
             Seed = executionSettings.Seed,
+#pragma warning restore OPENAI001 // Type is for evaluation purposes only and is subject to change or removal in future updates. Suppress this diagnostic to proceed.
             EndUserId = executionSettings.User,
             TopLogProbabilityCount = executionSettings.TopLogprobs,
             IncludeLogProbabilities = executionSettings.Logprobs,

dotnet/src/Connectors/Connectors.OpenAI.UnitTests/Core/ClientCoreTests.cs

@@ -67,9 +67,9 @@ public void ItUsesEndpointAsExpected(string? clientBaseAddress, string? provided
         var clientCore = new ClientCore("model", "apiKey", endpoint: endpoint, httpClient: client);
 
         // Assert
-        Assert.Equal(endpoint ?? client?.BaseAddress ?? new Uri("https://api.openai.com/"), clientCore.Endpoint);
+        Assert.Equal(endpoint ?? client?.BaseAddress ?? new Uri("https://api.openai.com/v1"), clientCore.Endpoint);
         Assert.True(clientCore.Attributes.ContainsKey(AIServiceExtensions.EndpointKey));
-        Assert.Equal(endpoint?.ToString() ?? client?.BaseAddress?.ToString() ?? "https://api.openai.com/", clientCore.Attributes[AIServiceExtensions.EndpointKey]);
+        Assert.Equal(endpoint?.ToString() ?? client?.BaseAddress?.ToString() ?? "https://api.openai.com/v1", clientCore.Attributes[AIServiceExtensions.EndpointKey]);
 
         client?.Dispose();
     }

dotnet/src/Connectors/Connectors.OpenAI.UnitTests/Services/OpenAIChatCompletionServiceTests.cs

@@ -76,10 +76,11 @@ public void ConstructorWithApiKeyWorksCorrectly(bool includeLoggerFactory)
     }
 
     [Theory]
-    [InlineData("http://localhost:1234/v1/chat/completions", "http://localhost:1234/v1/chat/completions")] // Uses full path when provided
-    [InlineData("http://localhost:1234/", "http://localhost:1234/v1/chat/completions")]
-    [InlineData("http://localhost:8080", "http://localhost:8080/v1/chat/completions")]
-    [InlineData("https://something:8080", "https://something:8080/v1/chat/completions")] // Accepts TLS Secured endpoints
+    [InlineData("http://localhost:1234", "http://localhost:1234/chat/completions")]
+    [InlineData("http://localhost:8080", "http://localhost:8080/chat/completions")]
+    [InlineData("https://something:8080", "https://something:8080/chat/completions")] // Accepts TLS Secured endpoints
+    [InlineData("http://localhost:1234/v2", "http://localhost:1234/v2/chat/completions")]
+    [InlineData("http://localhost:8080/v2", "http://localhost:8080/v2/chat/completions")]
     public async Task ItUsesCustomEndpointsWhenProvidedDirectlyAsync(string endpointProvided, string expectedEndpoint)
     {
         // Arrange
@@ -95,10 +96,11 @@ public async Task ItUsesCustomEndpointsWhenProvidedDirectlyAsync(string endpoint
     }
 
     [Theory]
-    [InlineData("http://localhost:1234/v1/chat/completions", "http://localhost:1234/v1/chat/completions")] // Uses full path when provided
-    [InlineData("http://localhost:1234/", "http://localhost:1234/v1/chat/completions")]
-    [InlineData("http://localhost:8080", "http://localhost:8080/v1/chat/completions")]
-    [InlineData("https://something:8080", "https://something:8080/v1/chat/completions")] // Accepts TLS Secured endpoints
+    [InlineData("http://localhost:1234", "http://localhost:1234/chat/completions")]
+    [InlineData("http://localhost:8080", "http://localhost:8080/chat/completions")]
+    [InlineData("https://something:8080", "https://something:8080/chat/completions")] // Accepts TLS Secured endpoints
+    [InlineData("http://localhost:1234/v2", "http://localhost:1234/v2/chat/completions")]
+    [InlineData("http://localhost:8080/v2", "http://localhost:8080/v2/chat/completions")]
     public async Task ItUsesCustomEndpointsWhenProvidedAsBaseAddressAsync(string endpointProvided, string expectedEndpoint)
     {
         // Arrange
@@ -127,7 +129,7 @@ public async Task ItUsesHttpClientEndpointIfProvidedEndpointIsMissingAsync()
         await chatCompletion.GetChatMessageContentsAsync(this._chatHistoryForTest, this._executionSettings);
 
         // Assert
-        Assert.Equal("http://localhost:12312/v1/chat/completions", this._messageHandlerStub.RequestUri!.ToString());
+        Assert.Equal("http://localhost:12312/chat/completions", this._messageHandlerStub.RequestUri!.ToString());
     }
 
     [Fact]

dotnet/src/Connectors/Connectors.OpenAI/Core/ClientCore.ChatCompletion.cs

@@ -650,7 +650,9 @@ protected virtual ChatCompletionOptions CreateChatCompletionOptions(
             TopP = (float?)executionSettings.TopP,
             FrequencyPenalty = (float?)executionSettings.FrequencyPenalty,
             PresencePenalty = (float?)executionSettings.PresencePenalty,
+#pragma warning disable OPENAI001 // Type is for evaluation purposes only and is subject to change or removal in future updates. Suppress this diagnostic to proceed.
             Seed = executionSettings.Seed,
+#pragma warning restore OPENAI001 // Type is for evaluation purposes only and is subject to change or removal in future updates. Suppress this diagnostic to proceed.
             EndUserId = executionSettings.User,
             TopLogProbabilityCount = executionSettings.TopLogprobs,
             IncludeLogProbabilities = executionSettings.Logprobs,

dotnet/src/Connectors/Connectors.OpenAI/Core/ClientCore.TextToAudio.cs

@@ -37,7 +37,7 @@ internal async Task<IReadOnlyList<AudioContent>> GetAudioContentsAsync(
         SpeechGenerationOptions options = new()
         {
             ResponseFormat = responseFormat,
-            Speed = audioExecutionSettings.Speed,
+            SpeedRatio = audioExecutionSettings.Speed,
         };
 
         ClientResult<BinaryData> response = await RunRequestAsync(() => this.Client!.GetAudioClient(targetModel).GenerateSpeechAsync(prompt, GetGeneratedSpeechVoice(audioExecutionSettings?.Voice), options, cancellationToken)).ConfigureAwait(false);
@@ -58,15 +58,17 @@ private static GeneratedSpeechVoice GetGeneratedSpeechVoice(string? voice)
         };
 
     private static (GeneratedSpeechFormat? Format, string? MimeType) GetGeneratedSpeechFormatAndMimeType(string? format)
-        => format?.ToUpperInvariant() switch
+    {
+        switch (format?.ToUpperInvariant())
         {
-            "WAV" => (GeneratedSpeechFormat.Wav, "audio/wav"),
-            "MP3" => (GeneratedSpeechFormat.Mp3, "audio/mpeg"),
-            "OPUS" => (GeneratedSpeechFormat.Opus, "audio/opus"),
-            "FLAC" => (GeneratedSpeechFormat.Flac, "audio/flac"),
-            "AAC" => (GeneratedSpeechFormat.Aac, "audio/aac"),
-            "PCM" => (GeneratedSpeechFormat.Pcm, "audio/l16"),
-            null => (null, null),
-            _ => throw new NotSupportedException($"The format '{format}' is not supported.")
-        };
+            case "WAV": return (GeneratedSpeechFormat.Wav, "audio/wav");
+            case "MP3": return (GeneratedSpeechFormat.Mp3, "audio/mpeg");
+            case "OPUS": return (GeneratedSpeechFormat.Opus, "audio/opus");
+            case "FLAC": return (GeneratedSpeechFormat.Flac, "audio/flac");
+            case "AAC": return (GeneratedSpeechFormat.Aac, "audio/aac");
+            case "PCM": return (GeneratedSpeechFormat.Pcm, "audio/l16");
+            case null: return (null, null);
+            default: throw new NotSupportedException($"The format '{format}' is not supported.");
+        }
+    }
 }

dotnet/src/Connectors/Connectors.OpenAI/Core/ClientCore.cs

@@ -36,7 +36,7 @@ internal partial class ClientCore
     /// <summary>
     /// Default OpenAI API endpoint.
     /// </summary>
-    private const string OpenAIEndpoint = "https://api.openai.com/";
+    private const string OpenAIV1Endpoint = "https://api.openai.com/v1";
 
     /// <summary>
     /// Identifier of the default model to use
@@ -104,7 +104,7 @@ internal ClientCore(
         if (this.Endpoint is null)
         {
             Verify.NotNullOrWhiteSpace(apiKey); // For Public OpenAI Endpoint a key must be provided.
-            this.Endpoint = new Uri(OpenAIEndpoint);
+            this.Endpoint = new Uri(OpenAIV1Endpoint);
         }
         else if (string.IsNullOrEmpty(apiKey))
         {

dotnet/src/Connectors/Connectors.OpenAI/Services/OpenAIChatCompletionService.cs

@@ -67,29 +67,11 @@ public OpenAIChatCompletionService(
             HttpClient? httpClient = null,
             ILoggerFactory? loggerFactory = null)
     {
-        Uri? internalClientEndpoint = null;
-        var providedEndpoint = endpoint ?? httpClient?.BaseAddress;
-        if (providedEndpoint is not null)
-        {
-            // As OpenAI Client automatically adds the chat completions endpoint, we remove it to avoid duplication.
-            const string PathAndQueryPattern = "v1/chat/completions";
-            var providedEndpointText = providedEndpoint.ToString();
-            int index = providedEndpointText.IndexOf(PathAndQueryPattern, StringComparison.OrdinalIgnoreCase);
-            if (index >= 0)
-            {
-                internalClientEndpoint = new Uri($"{providedEndpointText.Substring(0, index)}{providedEndpointText.Substring(index + PathAndQueryPattern.Length)}");
-            }
-            else
-            {
-                internalClientEndpoint = providedEndpoint;
-            }
-        }
-
         this._client = new(
             modelId,
             apiKey,
             organization,
-            internalClientEndpoint,
+            endpoint ?? httpClient?.BaseAddress,
             httpClient,
             loggerFactory?.CreateLogger(typeof(OpenAIChatCompletionService)));
     }