April 2024 Binary Update (#662)

* Updated binaries, using [this build](https://github.com/SciSharp/LLamaSharp/actions/runs/8654672719/job/23733195669) for llama.cpp commit `f7001ccc5aa359fcf41bba19d1c99c3d25c9bcc7`.

 - Added all new functions.
 - Moved some functions (e.g. `SafeLlamaModelHandle` specific functions) into `SafeLlamaModelHandle.cs`
 - Exposed tokens on `SafeLlamaModelHandle` and `LLamaWeights` through a `Tokens` property. As new special tokens are added in the future they can be added here.
 - Changed all token properties to return nullable tokens, to handle some models not having some tokens.
 - Fixed `DefaultSamplingPipeline` to handle no newline token in some models.

* Moved native methods to more specific locations.

 - Context specific things have been moved into `SafeLLamaContextHandle.cs` and made private - they're exposed through C# properties and methods already.
 - Checking that GPU layer count is zero if GPU offload is not supported.
 - Moved methods for creating default structs (`llama_model_quantize_default_params` and `llama_context_default_params`) into relevant structs.

* Removed exception if `GpuLayerCount > 0` when GPU is not supported.

* - Added low level wrapper methods for new per-sequence state load/save in `SafeLLamaContextHandle`
 - Added high level wrapper methods (save/load with `State` object or memory mapped file) in `LLamaContext`
 - Moved native methods for per-sequence state load/save into `SafeLLamaContextHandle`

* Added update and defrag methods for KV cache in `SafeLLamaContextHandle`

* Updated submodule to `f7001ccc5aa359fcf41bba19d1c99c3d25c9bcc7`

* Passing the sequence ID when saving a single sequence state

Author: martindevans

LLama.Examples/Examples/BatchedExecutorGuidance.cs

@@ -79,7 +79,7 @@ await AnsiConsole
                     guidance.Prompt(g);
 
                     // Early exit if we reach the natural end of the guided sentence
-                    if (g == model.EndOfSentenceToken)
+                    if (g == model.Tokens.EOS)
                         break;
 
                     // Update progress bar

LLama.Examples/Examples/GetEmbeddings.cs

@@ -9,7 +9,7 @@ public static void Run()
             string modelPath = UserSettings.GetModelPath();
 
             Console.ForegroundColor = ConsoleColor.DarkGray;
-            var @params = new ModelParams(modelPath) { EmbeddingMode = true };
+            var @params = new ModelParams(modelPath) { Embeddings = true };
             using var weights = LLamaWeights.LoadFromFile(@params);
             var embedder = new LLamaEmbedder(weights, @params);
 

LLama.Examples/Examples/SemanticKernelMemory.cs

@@ -20,7 +20,7 @@ public static async Task Run()
             var parameters = new ModelParams(modelPath)
             {
                 Seed = seed,
-                EmbeddingMode = true
+                Embeddings = true
             };
 
             using var model = LLamaWeights.LoadFromFile(parameters);

LLama.KernelMemory/BuilderExtensions.cs

@@ -84,7 +84,7 @@ public static IKernelMemoryBuilder WithLLamaSharpDefaults(this IKernelMemoryBuil
                 ContextSize = config?.ContextSize ?? 2048,
                 Seed = config?.Seed ?? 0,
                 GpuLayerCount = config?.GpuLayerCount ?? 20,
-                EmbeddingMode = true,
+                Embeddings = true,
                 MainGpu = config?.MainGpu ?? 0,
                 SplitMode = config?.SplitMode ?? GPUSplitMode.None,
             };

LLama.KernelMemory/LLamaSharpTextEmbeddingGenerator.cs

@@ -29,7 +29,7 @@ public LLamaSharpTextEmbeddingGenerator(LLamaSharpConfig config)
             this._config = config;
             var @params = new ModelParams(_config.ModelPath)
             {
-                EmbeddingMode = true,
+                Embeddings = true,
                 MainGpu = _config.MainGpu,
                 SplitMode = _config.SplitMode
             };
@@ -49,7 +49,7 @@ public LLamaSharpTextEmbeddingGenerator(LLamaSharpConfig config, LLamaWeights we
             this._config = config;
             var @params = new ModelParams(_config.ModelPath)
             {
-                EmbeddingMode = true,
+                Embeddings = true,
                 MainGpu = _config.MainGpu,
                 SplitMode = _config.SplitMode
             };

LLama.Unittest/BasicTest.cs

@@ -15,7 +15,7 @@ public sealed class BasicTest
         public BasicTest(ITestOutputHelper testOutputHelper)
         {
             _testOutputHelper = testOutputHelper;
-            _params = new ModelParams(Constants.ModelPath)
+            _params = new ModelParams(Constants.GenerativeModelPath)
             {
                 ContextSize = 2048
             };

LLama.Unittest/BeamTests.cs

@@ -15,7 +15,7 @@ public sealed class BeamTests
     public BeamTests(ITestOutputHelper testOutputHelper)
     {
         _testOutputHelper = testOutputHelper;
-        _params = new ModelParams(Constants.ModelPath)
+        _params = new ModelParams(Constants.GenerativeModelPath)
         {
             ContextSize = 2048
         };

LLama.Unittest/Constants.cs

@@ -2,9 +2,11 @@
 {
     internal static class Constants
     {
-        public static string ModelPath = "Models/llama-2-7b-chat.Q3_K_S.gguf";
-        public static string LLavaModelPath = "Models/llava-v1.6-mistral-7b.Q3_K_XS.gguf";
-        public static string LLavaMmpPath = "Models/mmproj-model-f16.gguf";
-        public static string LLavaImage = "Models/extreme-ironing-taxi-610x427.jpg";
+        public static readonly string GenerativeModelPath = "Models/llama-2-7b-chat.Q3_K_S.gguf";
+        public static readonly string EmbeddingModelPath = "Models/all-MiniLM-L12-v2.Q8_0.gguf";
+
+        public static readonly string LLavaModelPath = "Models/llava-v1.6-mistral-7b.Q3_K_XS.gguf";
+        public static readonly string LLavaMmpPath = "Models/mmproj-model-f16.gguf";
+        public static readonly string LLavaImage = "Models/extreme-ironing-taxi-610x427.jpg";
     }
 }

LLama.Unittest/GrammarTest.cs

@@ -12,7 +12,7 @@ public sealed class GrammarTest
 
         public GrammarTest()
         {
-            _params = new ModelParams(Constants.ModelPath)
+            _params = new ModelParams(Constants.GenerativeModelPath)
             {
                 ContextSize = 2048,
                 Seed = 92,

LLama.Unittest/LLama.Unittest.csproj

@@ -31,6 +31,9 @@
     <DownloadFile SourceUrl="https://huggingface.co/TheBloke/Llama-2-7b-Chat-GGUF/resolve/main/llama-2-7b-chat.Q3_K_S.gguf" DestinationFolder="Models" DestinationFileName="llama-2-7b-chat.Q3_K_S.gguf" SkipUnchangedFiles="true"></DownloadFile>
     <DownloadFile SourceUrl="https://huggingface.co/cjpais/llava-1.6-mistral-7b-gguf/resolve/main/llava-v1.6-mistral-7b.Q3_K_XS.gguf" DestinationFolder="Models" DestinationFileName="llava-v1.6-mistral-7b.Q3_K_XS.gguf" SkipUnchangedFiles="true"></DownloadFile>
     <DownloadFile SourceUrl="https://huggingface.co/cjpais/llava-1.6-mistral-7b-gguf/resolve/main/mmproj-model-f16.gguf" DestinationFolder="Models" DestinationFileName="mmproj-model-f16.gguf" SkipUnchangedFiles="true"></DownloadFile>
+    <DownloadFile SourceUrl="https://huggingface.co/leliuga/all-MiniLM-L12-v2-GGUF/resolve/main/all-MiniLM-L12-v2.Q8_0.gguf" DestinationFolder="Models" DestinationFileName="all-MiniLM-L12-v2.Q8_0.gguf" SkipUnchangedFiles="true"></DownloadFile>
+
+    
   </Target>
 
   <ItemGroup>
@@ -43,6 +46,9 @@
   </ItemGroup>
 
   <ItemGroup>
+    <None Update="Models\all-MiniLM-L12-v2.Q8_0.gguf">
+      <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
+    </None>
     <None Update="Models\llama-2-7b-chat.Q3_K_S.gguf">
       <CopyToOutputDirectory>PreserveNewest</CopyToOutputDirectory>
     </None>