chore: import upstream snapshot with attribution

2026-07-13 13:21:23 +08:00
commit b957a53def
5423 changed files with 863745 additions and 0 deletions
@@ -0,0 +1,106 @@
+// Copyright (c) Microsoft. All rights reserved.
+
+using Microsoft.SemanticKernel;
+using Microsoft.SemanticKernel.Connectors.HuggingFace;
+using xRetry;
+
+#pragma warning disable format // Format item can be simplified
+#pragma warning disable CA1861 // Avoid constant arrays as arguments
+
+namespace TextGeneration;
+
+// The following example shows how to use Semantic Kernel with HuggingFace API.
+public class HuggingFace_TextGeneration(ITestOutputHelper helper) : BaseTest(helper)
+{
+    private const string DefaultModel = "HuggingFaceH4/zephyr-7b-beta";
+
+    /// <summary>
+    /// This example uses HuggingFace Inference API to access hosted models.
+    /// More information here: <see href="https://huggingface.co/inference-api"/>
+    /// </summary>
+    [Fact]
+    public async Task RunInferenceApiExampleAsync()
+    {
+        Console.WriteLine("\n======== HuggingFace Inference API example ========\n");
+
+        Kernel kernel = Kernel.CreateBuilder()
+            .AddHuggingFaceTextGeneration(
+                model: TestConfiguration.HuggingFace.ModelId ?? DefaultModel,
+                apiKey: TestConfiguration.HuggingFace.ApiKey)
+            .Build();
+
+        var questionAnswerFunction = kernel.CreateFunctionFromPrompt("Question: {{$input}}; Answer:");
+
+        var result = await kernel.InvokeAsync(questionAnswerFunction, new() { ["input"] = "What is New York?" });
+
+        Console.WriteLine(result.GetValue<string>());
+    }
+
+    /// <summary>
+    /// Some Hugging Face models support streaming responses, configure using the HuggingFace ModelId setting.
+    /// </summary>
+    /// <remarks>
+    /// Tested with HuggingFaceH4/zephyr-7b-beta model.
+    /// </remarks>
+    [RetryFact(typeof(HttpOperationException))]
+    public async Task RunStreamingExampleAsync()
+    {
+        string model = TestConfiguration.HuggingFace.ModelId ?? DefaultModel;
+
+        Console.WriteLine($"\n======== HuggingFace {model} streaming example ========\n");
+
+        Kernel kernel = Kernel.CreateBuilder()
+            .AddHuggingFaceTextGeneration(
+                model: model,
+                apiKey: TestConfiguration.HuggingFace.ApiKey)
+            .Build();
+
+        var settings = new HuggingFacePromptExecutionSettings { UseCache = false };
+
+        var questionAnswerFunction = kernel.CreateFunctionFromPrompt("Question: {{$input}}; Answer:", new HuggingFacePromptExecutionSettings
+        {
+            UseCache = false
+        });
+
+        await foreach (string text in kernel.InvokePromptStreamingAsync<string>("Question: {{$input}}; Answer:", new(settings) { ["input"] = "What is New York?" }))
+        {
+            Console.Write(text);
+        }
+    }
+
+    /// <summary>
+    /// This example uses HuggingFace Llama 2 model and local HTTP server from Semantic Kernel repository.
+    /// How to setup local HTTP server: <see href="https://github.com/microsoft/semantic-kernel/blob/main/samples/apps/hugging-face-http-server/README.md"/>.
+    /// <remarks>
+    /// Additional access is required to download Llama 2 model and run it locally.
+    /// How to get access:
+    /// 1. Visit <see href="https://ai.meta.com/resources/models-and-libraries/llama-downloads/"/> and complete request access form.
+    /// 2. Visit <see href="https://huggingface.co/meta-llama/Llama-2-7b-hf"/> and complete form "Access Llama 2 on Hugging Face".
+    /// Note: Your Hugging Face account email address MUST match the email you provide on the Meta website, or your request will not be approved.
+    /// </remarks>
+    /// </summary>
+    [Fact(Skip = "Requires local model or Huggingface Pro subscription")]
+    public async Task RunLlamaExampleAsync()
+    {
+        Console.WriteLine("\n======== HuggingFace Llama 2 example ========\n");
+
+        // HuggingFace Llama 2 model: https://huggingface.co/meta-llama/Llama-2-7b-hf
+        const string Model = "meta-llama/Llama-2-7b-hf";
+
+        // HuggingFace local HTTP server endpoint
+        // const string Endpoint = "http://localhost:5000/completions";
+
+        Kernel kernel = Kernel.CreateBuilder()
+            .AddHuggingFaceTextGeneration(
+                model: Model,
+                //endpoint: Endpoint,
+                apiKey: TestConfiguration.HuggingFace.ApiKey)
+            .Build();
+
+        var questionAnswerFunction = kernel.CreateFunctionFromPrompt("Question: {{$input}}; Answer:");
+
+        var result = await kernel.InvokeAsync(questionAnswerFunction, new() { ["input"] = "What is New York?" });
+
+        Console.WriteLine(result.GetValue<string>());
+    }
+}