🔧 Refactor chat completion implementation

- Refactored the chat completion implementation in `LLamaSharpChatCompletion.cs` to use `StatelessExecutor` instead of `InteractiveExecutor`.
- Updated the chat history prompt in `LLamaSharpChatCompletion.cs` to include a conversation between the assistant and the user.
- Modified the `HistoryTransform` class in `HistoryTransform.cs` to append the assistant role to the chat history prompt.
- Updated the constructor of `LLamaSharpChatCompletion` to accept optional parameters for `historyTransform` and `outputTransform`.
- Modified the `GetChatCompletionsAsync` and `GetChatCompletions` methods in `LLamaSharpChatCompletion.cs` to use the new `StatelessExecutor` and `outputTransform`.
- Updated the `ExtensionMethods.cs` file to include the assistant and system roles in the list of anti-prompts.

LLama.Examples/Examples/SemanticKernelChat.cs

@@ -16,12 +16,11 @@ namespace LLama.Examples.Examples
             // Load weights into memory
             var parameters = new ModelParams(modelPath);
             using var model = LLamaWeights.LoadFromFile(parameters);
-            using var context = model.CreateContext(parameters);
-            var ex = new InteractiveExecutor(context);
+            var ex = new StatelessExecutor(model, parameters);
 
             var chatGPT = new LLamaSharpChatCompletion(ex);
 
-            var chatHistory = chatGPT.CreateNewChat("You are a librarian, expert about books");
+            var chatHistory = chatGPT.CreateNewChat("This is a conversation between the assistant and the user. \n\n You are a librarian, expert about books. ");
 
             Console.WriteLine("Chat content:");
             Console.WriteLine("------------------------");

LLama.SemanticKernel/ChatCompletion/HistoryTransform.cs

@@ -1,4 +1,6 @@
-using static LLama.LLamaTransforms;
+using LLama.Common;
+using System.Text;
+using static LLama.LLamaTransforms;
 
 namespace LLamaSharp.SemanticKernel.ChatCompletion;
 
@@ -10,8 +12,6 @@ public class HistoryTransform : DefaultHistoryTransform
     /// <inheritdoc/>
     public override string HistoryToText(global::LLama.Common.ChatHistory history)
     {
-        var prompt = base.HistoryToText(history);
-        return prompt + "\nAssistant:";
-
+        return base.HistoryToText(history) + $"{AuthorRole.Assistant}: ";
     }
 }

LLama.SemanticKernel/ChatCompletion/LLamaSharpChatCompletion.cs

@@ -1,7 +1,9 @@
 using LLama;
+using LLama.Abstractions;
 using Microsoft.SemanticKernel.AI;
 using Microsoft.SemanticKernel.AI.ChatCompletion;
 using System.Runtime.CompilerServices;
+using static LLama.LLamaTransforms;
 
 namespace LLamaSharp.SemanticKernel.ChatCompletion;
 
@@ -10,10 +12,10 @@ namespace LLamaSharp.SemanticKernel.ChatCompletion;
 /// </summary>
 public sealed class LLamaSharpChatCompletion : IChatCompletion
 {
-    private const string UserRole = "user:";
-    private const string AssistantRole = "assistant:";
-    private ChatSession session;
+    private readonly StatelessExecutor _model;
     private ChatRequestSettings defaultRequestSettings;
+    private readonly IHistoryTransform historyTransform;
+    private readonly ITextStreamTransform outputTransform;
 
     private readonly Dictionary<string, string> _attributes = new();
 
@@ -30,18 +32,17 @@ public sealed class LLamaSharpChatCompletion : IChatCompletion
         };
     }
 
-    public LLamaSharpChatCompletion(InteractiveExecutor model, ChatRequestSettings? defaultRequestSettings = default)
+    public LLamaSharpChatCompletion(StatelessExecutor model,
+        ChatRequestSettings? defaultRequestSettings = default,
+        IHistoryTransform? historyTransform = null,
+        ITextStreamTransform? outputTransform = null)
     {
-        this.session = new ChatSession(model)
-            .WithHistoryTransform(new HistoryTransform())
-            .WithOutputTransform(new LLamaTransforms.KeywordTextOutputStreamTransform(new string[] { UserRole, AssistantRole }));
-        this.defaultRequestSettings = defaultRequestSettings ??= GetDefaultSettings();
-    }
-
-    public LLamaSharpChatCompletion(ChatSession session, ChatRequestSettings? defaultRequestSettings = default)
-    {
-        this.session = session;
-        this.defaultRequestSettings = defaultRequestSettings ??= GetDefaultSettings();
+        this._model = model;
+        this.defaultRequestSettings = defaultRequestSettings ?? GetDefaultSettings();
+        this.historyTransform = historyTransform ?? new HistoryTransform();
+        this.outputTransform = outputTransform ?? new KeywordTextOutputStreamTransform(new[] { $"{LLama.Common.AuthorRole.User}:",
+                                                                                            $"{LLama.Common.AuthorRole.Assistant}:",
+                                                                                            $"{LLama.Common.AuthorRole.System}:"});
     }
 
     /// <inheritdoc/>
@@ -60,14 +61,14 @@ public sealed class LLamaSharpChatCompletion : IChatCompletion
     /// <inheritdoc/>
     public Task<IReadOnlyList<IChatResult>> GetChatCompletionsAsync(ChatHistory chat, AIRequestSettings? requestSettings = null, CancellationToken cancellationToken = default)
     {
-        var settings = requestSettings != null 
+        var settings = requestSettings != null
             ? ChatRequestSettings.FromRequestSettings(requestSettings)
             : defaultRequestSettings;
+        var prompt = historyTransform.HistoryToText(chat.ToLLamaSharpChatHistory());
 
-        // This call is not awaited because LLamaSharpChatResult accepts an IAsyncEnumerable.
-        var result = this.session.ChatAsync(chat.ToLLamaSharpChatHistory(), settings.ToLLamaSharpInferenceParams(), cancellationToken);
+        var result = _model.InferAsync(prompt, settings.ToLLamaSharpInferenceParams(), cancellationToken);
 
-        return Task.FromResult<IReadOnlyList<IChatResult>>(new List<IChatResult> { new LLamaSharpChatResult(result) }.AsReadOnly());
+        return Task.FromResult<IReadOnlyList<IChatResult>>(new List<IChatResult> { new LLamaSharpChatResult(outputTransform.TransformAsync(result)) }.AsReadOnly());
     }
 
     /// <inheritdoc/>
@@ -78,10 +79,10 @@ public sealed class LLamaSharpChatCompletion : IChatCompletion
         var settings = requestSettings != null
             ? ChatRequestSettings.FromRequestSettings(requestSettings)
             : defaultRequestSettings;
-
+        var prompt = historyTransform.HistoryToText(chat.ToLLamaSharpChatHistory());
         // This call is not awaited because LLamaSharpChatResult accepts an IAsyncEnumerable.
-        var result = this.session.ChatAsync(chat.ToLLamaSharpChatHistory(), settings.ToLLamaSharpInferenceParams(), cancellationToken);
+        var result = _model.InferAsync(prompt, settings.ToLLamaSharpInferenceParams(), cancellationToken);
 
-        yield return new LLamaSharpChatResult(result);
+        yield return new LLamaSharpChatResult(outputTransform.TransformAsync(result));
     }
 }

LLama.SemanticKernel/ExtensionMethods.cs

@@ -35,7 +35,10 @@ public static class ExtensionMethods
             throw new ArgumentNullException(nameof(requestSettings));
         }
 
-        var antiPrompts = new List<string>(requestSettings.StopSequences) { AuthorRole.User.ToString() + ":" };
+        var antiPrompts = new List<string>(requestSettings.StopSequences)
+                                  { LLama.Common.AuthorRole.User.ToString() + ":" ,
+                                    LLama.Common.AuthorRole.Assistant.ToString() + ":",
+                                    LLama.Common.AuthorRole.System.ToString() + ":"};
         return new global::LLama.Common.InferenceParams
         {
             Temperature = (float)requestSettings.Temperature,