Code cleanup driven by R# suggestions:

- Made `NativeApi` into a `static class` (it's not intended to be instantiated)
 - Moved `LLamaTokenType` enum out into a separate file
 - Made `LLamaSeqId` and `LLamaPos` into `record struct`, convenient to have equality etc

LLama/Abstractions/IModelParams.cs

@@ -214,7 +214,7 @@ namespace LLama.Abstractions
         /// <summary>
         /// Get the key being overriden by this override
         /// </summary>
-        public string Key { get; init; }
+        public string Key { get; }
 
         internal LLamaModelKvOverrideType Type { get; }
 

LLama/Common/ChatHistory.cs

@@ -1,5 +1,4 @@
 using System.Collections.Generic;
-using System.IO;
 using System.Text.Json;
 using System.Text.Json.Serialization;
 
@@ -37,6 +36,7 @@ namespace LLama.Common
     /// </summary>
     public class ChatHistory
     {
+        private static readonly JsonSerializerOptions _jsonOptions = new() { WriteIndented = true };
 
         /// <summary>
         /// Chat message representation
@@ -96,12 +96,7 @@ namespace LLama.Common
         /// <returns></returns>
         public string ToJson()
         {
-            return JsonSerializer.Serialize(
-                this,
-                new JsonSerializerOptions()
-                {
-                    WriteIndented = true
-                });
+            return JsonSerializer.Serialize(this, _jsonOptions);
         }
 
         /// <summary>

LLama/Common/FixedSizeQueue.cs

@@ -2,7 +2,6 @@
 using System.Collections;
 using System.Collections.Generic;
 using System.Linq;
-using LLama.Extensions;
 
 namespace LLama.Common
 {

LLama/Common/InferenceParams.cs

@@ -18,11 +18,13 @@ namespace LLama.Common
         /// number of tokens to keep from initial prompt
         /// </summary>
         public int TokensKeep { get; set; } = 0;
+
         /// <summary>
         /// how many new tokens to predict (n_predict), set to -1 to inifinitely generate response
         /// until it complete.
         /// </summary>
         public int MaxTokens { get; set; } = -1;
+
         /// <summary>
         /// logit bias for specific tokens
         /// </summary>

LLama/Extensions/DictionaryExtensions.cs

@@ -15,6 +15,7 @@ namespace LLama.Extensions
 
         internal static TValue GetValueOrDefaultImpl<TKey, TValue>(IReadOnlyDictionary<TKey, TValue> dictionary, TKey key, TValue defaultValue)
         {
+            // ReSharper disable once CanSimplifyDictionaryTryGetValueWithGetValueOrDefault (this is a shim for  that method!)
             return dictionary.TryGetValue(key, out var value) ? value : defaultValue;
         }
     }

LLama/Grammars/Grammar.cs

@@ -15,7 +15,7 @@ namespace LLama.Grammars
         /// <summary>
         /// Index of the initial rule to start from
         /// </summary>
-        public ulong StartRuleIndex { get; set; }
+        public ulong StartRuleIndex { get; }
 
         /// <summary>
         /// The rules which make up this grammar
@@ -121,6 +121,12 @@ namespace LLama.Grammars
                         case LLamaGrammarElementType.CHAR_ALT:
                         case LLamaGrammarElementType.CHAR_RNG_UPPER:
                             break;
+
+                        case LLamaGrammarElementType.END:
+                        case LLamaGrammarElementType.ALT:
+                        case LLamaGrammarElementType.RULE_REF:
+                        case LLamaGrammarElementType.CHAR:
+                        case LLamaGrammarElementType.CHAR_NOT:
                         default:
                             output.Append("] ");
                             break;

LLama/LLamaContext.cs

@@ -43,7 +43,7 @@ namespace LLama
         /// <summary>
         /// The context params set for this context
         /// </summary>
-        public IContextParams Params { get; set; }
+        public IContextParams Params { get; }
 
         /// <summary>
         /// The native handle, which is used to be passed to the native APIs
@@ -56,15 +56,6 @@ namespace LLama
         /// </summary>
         public Encoding Encoding { get; }
 
-        internal LLamaContext(SafeLLamaContextHandle nativeContext, IContextParams @params, ILogger? logger = null)
-        {
-            Params = @params;
-
-            _logger = logger;
-            Encoding = @params.Encoding;
-            NativeHandle = nativeContext;
-        }
-
         /// <summary>
         /// Create a new LLamaContext for the given LLamaWeights
         /// </summary>

LLama/LLamaEmbedder.cs

@@ -12,17 +12,15 @@ namespace LLama
     public sealed class LLamaEmbedder
         : IDisposable
     {
-        private readonly LLamaContext _ctx;
-
         /// <summary>
         /// Dimension of embedding vectors
         /// </summary>
-        public int EmbeddingSize => _ctx.EmbeddingSize;
+        public int EmbeddingSize => Context.EmbeddingSize;
 
         /// <summary>
         /// LLama Context
         /// </summary>
-        public LLamaContext Context => this._ctx;
+        public LLamaContext Context { get; }
 
         /// <summary>
         /// Create a new embedder, using the given LLamaWeights
@@ -33,7 +31,7 @@ namespace LLama
         public LLamaEmbedder(LLamaWeights weights, IContextParams @params, ILogger? logger = null)
         {
             @params.EmbeddingMode = true;
-            _ctx = weights.CreateContext(@params, logger);
+            Context = weights.CreateContext(@params, logger);
         }
 
         /// <summary>
@@ -72,20 +70,20 @@ namespace LLama
         /// <exception cref="RuntimeError"></exception>
         public float[] GetEmbeddings(string text, bool addBos)
         {
-            var embed_inp_array = _ctx.Tokenize(text, addBos);
+            var embed_inp_array = Context.Tokenize(text, addBos);
 
             // TODO(Rinne): deal with log of prompt
 
             if (embed_inp_array.Length > 0)
-                _ctx.Eval(embed_inp_array, 0);
+                Context.Eval(embed_inp_array, 0);
 
             unsafe
             {
-                var embeddings = NativeApi.llama_get_embeddings(_ctx.NativeHandle);
+                var embeddings = NativeApi.llama_get_embeddings(Context.NativeHandle);
                 if (embeddings == null)
                     return Array.Empty<float>();
 
-                return new Span<float>(embeddings, EmbeddingSize).ToArray();
+                return embeddings.ToArray();
             }
         }
 
@@ -94,7 +92,7 @@ namespace LLama
         /// </summary>
         public void Dispose()
         {
-            _ctx.Dispose();
+            Context.Dispose();
         }
 
     }

LLama/LLamaWeights.cs

@@ -64,7 +64,7 @@ namespace LLama
         /// </summary>
         public IReadOnlyDictionary<string, string> Metadata { get; set; }
 
-        internal LLamaWeights(SafeLlamaModelHandle weights)
+        private LLamaWeights(SafeLlamaModelHandle weights)
         {
             NativeHandle = weights;
             Metadata = weights.ReadMetadata();

LLama/Native/LLamaKvCacheView.cs

@@ -14,7 +14,7 @@ public struct LLamaKvCacheViewCell
     /// May be negative if the cell is not populated.
     /// </summary>
     public LLamaPos pos;
-};
+}
 
 /// <summary>
 /// An updateable view of the KV cache (llama_kv_cache_view)
@@ -130,7 +130,7 @@ public class LLamaKvCacheViewSafeHandle
     }
 }
 
-partial class NativeApi
+public static partial class NativeApi
 {
     /// <summary>
     /// Create an empty KV cache view. (use only for debugging purposes)

LLama/Native/LLamaPos.cs

@@ -6,7 +6,7 @@ namespace LLama.Native;
 /// Indicates position in a sequence
 /// </summary>
 [StructLayout(LayoutKind.Sequential)]
-public struct LLamaPos
+public record struct LLamaPos
 {
     /// <summary>
     /// The raw value
@@ -17,7 +17,7 @@ public struct LLamaPos
     /// Create a new LLamaPos
     /// </summary>
     /// <param name="value"></param>
-    public LLamaPos(int value)
+    private LLamaPos(int value)
     {
         Value = value;
     }

LLama/Native/LLamaSeqId.cs

@@ -6,7 +6,7 @@ namespace LLama.Native;
 /// ID for a sequence in a batch
 /// </summary>
 [StructLayout(LayoutKind.Sequential)]
-public struct LLamaSeqId
+public record struct LLamaSeqId
 {
     /// <summary>
     /// The raw value
@@ -17,7 +17,7 @@ public struct LLamaSeqId
     /// Create a new LLamaSeqId 
     /// </summary>
     /// <param name="value"></param>
-    public LLamaSeqId(int value)
+    private LLamaSeqId(int value)
     {
         Value = value;
     }

LLama/Native/LLamaTokenType.cs

@@ -0,0 +1,12 @@
+namespace LLama.Native;
+
+public enum LLamaTokenType
+{
+    LLAMA_TOKEN_TYPE_UNDEFINED = 0,
+    LLAMA_TOKEN_TYPE_NORMAL = 1,
+    LLAMA_TOKEN_TYPE_UNKNOWN = 2,
+    LLAMA_TOKEN_TYPE_CONTROL = 3,
+    LLAMA_TOKEN_TYPE_USER_DEFINED = 4,
+    LLAMA_TOKEN_TYPE_UNUSED = 5,
+    LLAMA_TOKEN_TYPE_BYTE = 6,
+}

LLama/Native/NativeApi.BeamSearch.cs

@@ -3,7 +3,7 @@ using System.Runtime.InteropServices;
 
 namespace LLama.Native;
 
-public partial class NativeApi
+public static partial class NativeApi
 {
     /// <summary>
     /// Type of pointer to the beam_search_callback function.

LLama/Native/NativeApi.Grammar.cs

@@ -5,7 +5,7 @@ namespace LLama.Native
 {
     using llama_token = Int32;
 
-	public unsafe partial class NativeApi
+	public static partial class NativeApi
     {
 		/// <summary>
 		/// Create a new grammar from the given set of grammar rules
@@ -15,7 +15,7 @@ namespace LLama.Native
 		/// <param name="start_rule_index"></param>
 		/// <returns></returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern IntPtr llama_grammar_init(LLamaGrammarElement** rules, ulong n_rules, ulong start_rule_index);
+        public static extern unsafe IntPtr llama_grammar_init(LLamaGrammarElement** rules, ulong n_rules, ulong start_rule_index);
 
         /// <summary>
         /// Free all memory from the given SafeLLamaGrammarHandle

LLama/Native/NativeApi.Load.cs

@@ -4,13 +4,12 @@ using System;
 using System.Collections.Generic;
 using System.Diagnostics;
 using System.IO;
-using System.Linq;
 using System.Runtime.InteropServices;
 using System.Text.Json;
 
 namespace LLama.Native
 {
-    public partial class NativeApi
+    public static partial class NativeApi
     {
         static NativeApi()
         {
@@ -97,22 +96,13 @@ namespace LLama.Native
             }
 
             if (string.IsNullOrEmpty(version))
-            {
                 return -1;
-            }
-            else
-            {
-                version = version.Split('.')[0];
-                bool success = int.TryParse(version, out var majorVersion);
-                if (success)
-                {
-                    return majorVersion;
-                }
-                else
-                {
-                    return -1;
-                }
-            }
+
+            version = version.Split('.')[0];
+            if (int.TryParse(version, out var majorVersion))
+                return majorVersion;
+
+            return -1;
         }
 
         private static string GetCudaVersionFromPath(string cudaPath)
@@ -129,7 +119,7 @@ namespace LLama.Native
                     {
                         return string.Empty;
                     }
-                    return versionNode.GetString();
+                    return versionNode.GetString() ?? "";
                 }
             }
             catch (Exception)
@@ -169,18 +159,14 @@ namespace LLama.Native
             {
                 platform = OSPlatform.OSX;
                 suffix = ".dylib";
-                if (System.Runtime.Intrinsics.Arm.ArmBase.Arm64.IsSupported)
-                {
-                    prefix = "runtimes/osx-arm64/native/";
-                }
-                else
-                {
-                    prefix = "runtimes/osx-x64/native/";
-                }
+
+                prefix = System.Runtime.Intrinsics.Arm.ArmBase.Arm64.IsSupported
+                       ? "runtimes/osx-arm64/native/"
+                       : "runtimes/osx-x64/native/";
             }
             else
             {
-                throw new RuntimeError($"Your system plarform is not supported, please open an issue in LLamaSharp.");
+                throw new RuntimeError("Your system plarform is not supported, please open an issue in LLamaSharp.");
             }
             Log($"Detected OS Platform: {platform}", LogLevel.Information);
 
@@ -275,15 +261,15 @@ namespace LLama.Native
 
             var libraryTryLoadOrder = GetLibraryTryOrder(configuration);
 
-            string[] preferredPaths = configuration.SearchDirectories;
-            string[] possiblePathPrefix = new string[] {
-                System.AppDomain.CurrentDomain.BaseDirectory,
+            var preferredPaths = configuration.SearchDirectories;
+            var possiblePathPrefix = new[] {
+                AppDomain.CurrentDomain.BaseDirectory,
                 Path.GetDirectoryName(System.Reflection.Assembly.GetExecutingAssembly().Location) ?? ""
             };
 
-            var tryFindPath = (string filename) =>
+            string TryFindPath(string filename)
             {
-                foreach(var path in preferredPaths)
+                foreach (var path in preferredPaths)
                 {
                     if (File.Exists(Path.Combine(path, filename)))
                     {
@@ -291,7 +277,7 @@ namespace LLama.Native
                     }
                 }
 
-                foreach(var path in possiblePathPrefix)
+                foreach (var path in possiblePathPrefix)
                 {
                     if (File.Exists(Path.Combine(path, filename)))
                     {
@@ -300,21 +286,19 @@ namespace LLama.Native
                 }
 
                 return filename;
-            };
+            }
 
             foreach (var libraryPath in libraryTryLoadOrder)
             {
-                var fullPath = tryFindPath(libraryPath);
+                var fullPath = TryFindPath(libraryPath);
                 var result = TryLoad(fullPath, true);
                 if (result is not null && result != IntPtr.Zero)
                 {
                     Log($"{fullPath} is selected and loaded successfully.", LogLevel.Information);
-                    return result ?? IntPtr.Zero;
-                }
-                else
-                {
-                    Log($"Tried to load {fullPath} but failed.", LogLevel.Information);
+                    return (IntPtr)result;
                 }
+
+                Log($"Tried to load {fullPath} but failed.", LogLevel.Information);
             }
 
             if (!configuration.AllowFallback)

LLama/Native/NativeApi.Quantize.cs

@@ -2,7 +2,7 @@
 
 namespace LLama.Native
 {
-    public partial class NativeApi
+    public static partial class NativeApi
     {
         /// <summary>
         /// Returns 0 on success

LLama/Native/NativeApi.Sampling.cs

@@ -5,7 +5,7 @@ namespace LLama.Native
 {
     using llama_token = Int32;
 
-    public unsafe partial class NativeApi
+    public static partial class NativeApi
     {
         /// <summary>
         /// Repetition penalty described in CTRL academic paper https://arxiv.org/abs/1909.05858, with negative logit fix.
@@ -19,7 +19,7 @@ namespace LLama.Native
         /// <param name="penalty_freq">Frequency and presence penalties described in OpenAI API https://platform.openai.com/docs/api-reference/parameter-details.</param>
         /// <param name="penalty_present">Frequency and presence penalties described in OpenAI API https://platform.openai.com/docs/api-reference/parameter-details.</param>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern void llama_sample_repetition_penalties(SafeLLamaContextHandle ctx,
+        public static extern unsafe void llama_sample_repetition_penalties(SafeLLamaContextHandle ctx,
                                                                     ref LLamaTokenDataArrayNative candidates,
                                                                     llama_token* last_tokens, ulong last_tokens_size,
                                                                     float penalty_repeat,

LLama/Native/NativeApi.cs

@@ -9,17 +9,6 @@ namespace LLama.Native
 {
     using llama_token = Int32;
 
-    public enum LLamaTokenType
-    {
-        LLAMA_TOKEN_TYPE_UNDEFINED = 0,
-        LLAMA_TOKEN_TYPE_NORMAL = 1,
-        LLAMA_TOKEN_TYPE_UNKNOWN = 2,
-        LLAMA_TOKEN_TYPE_CONTROL = 3,
-        LLAMA_TOKEN_TYPE_USER_DEFINED = 4,
-        LLAMA_TOKEN_TYPE_UNUSED = 5,
-        LLAMA_TOKEN_TYPE_BYTE = 6,
-    }
-
     /// <summary>
     /// Callback from llama.cpp with log messages
     /// </summary>
@@ -30,7 +19,7 @@ namespace LLama.Native
     /// <summary>
     /// Direct translation of the llama.cpp API
     /// </summary>
-	public unsafe partial class NativeApi
+	public static partial class NativeApi
     {
         /// <summary>
         /// A method that does nothing. This is a native method, calling it will force the llama native dependencies to be loaded.
@@ -165,7 +154,7 @@ namespace LLama.Native
         /// <param name="dest"></param>
         /// <returns>the number of bytes copied</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern ulong llama_copy_state_data(SafeLLamaContextHandle ctx, byte* dest);
+        public static extern unsafe ulong llama_copy_state_data(SafeLLamaContextHandle ctx, byte* dest);
 
         /// <summary>
         /// Set the state reading from the specified address
@@ -174,7 +163,7 @@ namespace LLama.Native
         /// <param name="src"></param>
         /// <returns>the number of bytes read</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern ulong llama_set_state_data(SafeLLamaContextHandle ctx, byte* src);
+        public static extern unsafe ulong llama_set_state_data(SafeLLamaContextHandle ctx, byte* src);
 
         /// <summary>
         /// Load session file
@@ -186,7 +175,7 @@ namespace LLama.Native
         /// <param name="n_token_count_out"></param>
         /// <returns></returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern bool llama_load_session_file(SafeLLamaContextHandle ctx, string path_session, llama_token[] tokens_out, ulong n_token_capacity, ulong* n_token_count_out);
+        public static extern unsafe bool llama_load_session_file(SafeLLamaContextHandle ctx, string path_session, llama_token[] tokens_out, ulong n_token_capacity, ulong* n_token_count_out);
 
         /// <summary>
         /// Save session file
@@ -211,7 +200,7 @@ namespace LLama.Native
         /// <returns>Returns 0 on success</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
         [Obsolete("use llama_decode() instead")]
-        public static extern int llama_eval(SafeLLamaContextHandle ctx, llama_token* tokens, int n_tokens, int n_past);
+        public static extern unsafe int llama_eval(SafeLLamaContextHandle ctx, llama_token* tokens, int n_tokens, int n_past);
 
         /// <summary>
         /// Convert the provided text into tokens.
@@ -228,34 +217,37 @@ namespace LLama.Native
         /// </returns>
         public static int llama_tokenize(SafeLLamaContextHandle ctx, string text, Encoding encoding, llama_token[] tokens, int n_max_tokens, bool add_bos, bool special)
         {
-            // Calculate number of bytes in text and borrow an array that large (+1 for nul byte)
-            var byteCount = encoding.GetByteCount(text);
-            var array = ArrayPool<byte>.Shared.Rent(byteCount + 1);
-            try
+            unsafe
             {
-                // Convert to bytes
-                fixed (char* textPtr = text)
-                fixed (byte* arrayPtr = array)
+                // Calculate number of bytes in text and borrow an array that large (+1 for nul byte)
+                var byteCount = encoding.GetByteCount(text);
+                var array = ArrayPool<byte>.Shared.Rent(byteCount + 1);
+                try
                 {
-                    encoding.GetBytes(textPtr, text.Length, arrayPtr, array.Length);
+                    // Convert to bytes
+                    fixed (char* textPtr = text)
+                    fixed (byte* arrayPtr = array)
+                    {
+                        encoding.GetBytes(textPtr, text.Length, arrayPtr, array.Length);
+                    }
+
+                    // Add a zero byte to the end to terminate the string
+                    array[byteCount] = 0;
+
+                    // Do the actual tokenization
+                    fixed (byte* arrayPtr = array)
+                    fixed (llama_token* tokensPtr = tokens)
+                        return llama_tokenize(ctx.ModelHandle, arrayPtr, byteCount, tokensPtr, n_max_tokens, add_bos, special);
+                }
+                finally
+                {
+                    ArrayPool<byte>.Shared.Return(array);
                 }
-
-                // Add a zero byte to the end to terminate the string
-                array[byteCount] = 0;
-
-                // Do the actual tokenization
-                fixed (byte* arrayPtr = array)
-                fixed (llama_token* tokensPtr = tokens)
-                    return llama_tokenize(ctx.ModelHandle, arrayPtr, byteCount, tokensPtr, n_max_tokens, add_bos, special);
-            }
-            finally
-            {
-                ArrayPool<byte>.Shared.Return(array);
             }
         }
 
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern byte* llama_token_get_text(SafeLlamaModelHandle model, llama_token token);
+        public static extern unsafe byte* llama_token_get_text(SafeLlamaModelHandle model, llama_token token);
 
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
         public static extern float llama_token_get_score(SafeLlamaModelHandle model, llama_token token);
@@ -281,7 +273,7 @@ namespace LLama.Native
         /// <param name="ctx"></param>
         /// <returns></returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern float* llama_get_logits(SafeLLamaContextHandle ctx);
+        public static extern unsafe float* llama_get_logits(SafeLLamaContextHandle ctx);
 
         /// <summary>
         /// Logits for the ith token. Equivalent to: llama_get_logits(ctx) + i*n_vocab
@@ -290,16 +282,24 @@ namespace LLama.Native
         /// <param name="i"></param>
         /// <returns></returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern float* llama_get_logits_ith(SafeLLamaContextHandle ctx, int i);
+        public static extern unsafe float* llama_get_logits_ith(SafeLLamaContextHandle ctx, int i);
 
         /// <summary>
         /// Get the embeddings for the input
-        /// shape: [n_embd] (1-dimensional)
         /// </summary>
         /// <param name="ctx"></param>
         /// <returns></returns>
-        [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern float* llama_get_embeddings(SafeLLamaContextHandle ctx);
+        public static Span<float> llama_get_embeddings(SafeLLamaContextHandle ctx)
+        {
+            unsafe
+            {
+                var ptr = llama_get_embeddings_native(ctx);
+                return new Span<float>(ptr, ctx.EmbeddingSize);
+            }
+
+            [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl, EntryPoint = "llama_get_embeddings")]
+            static extern unsafe float* llama_get_embeddings_native(SafeLLamaContextHandle ctx);
+        }
 
         /// <summary>
         /// Get the "Beginning of sentence" token
@@ -426,7 +426,7 @@ namespace LLama.Native
         /// <param name="buf_size"></param>
         /// <returns>The length of the string on success, or -1 on failure</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern int llama_model_meta_val_str(SafeLlamaModelHandle model, byte* key, byte* buf, long buf_size);
+        public static extern unsafe int llama_model_meta_val_str(SafeLlamaModelHandle model, byte* key, byte* buf, long buf_size);
 
         /// <summary>
         /// Get the number of metadata key/value pairs
@@ -445,7 +445,7 @@ namespace LLama.Native
         /// <param name="buf_size"></param>
         /// <returns>The length of the string on success, or -1 on failure</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern int llama_model_meta_key_by_index(SafeLlamaModelHandle model, int index, byte* buf, long buf_size);
+        public static extern unsafe int llama_model_meta_key_by_index(SafeLlamaModelHandle model, int index, byte* buf, long buf_size);
 
         /// <summary>
         /// Get metadata value as a string by index
@@ -456,7 +456,7 @@ namespace LLama.Native
         /// <param name="buf_size"></param>
         /// <returns>The length of the string on success, or -1 on failure</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern int llama_model_meta_val_str_by_index(SafeLlamaModelHandle model, int index, byte* buf, long buf_size);
+        public static extern unsafe int llama_model_meta_val_str_by_index(SafeLlamaModelHandle model, int index, byte* buf, long buf_size);
 
         /// <summary>
         /// Get a string describing the model type
@@ -466,7 +466,7 @@ namespace LLama.Native
         /// <param name="buf_size"></param>
         /// <returns>The length of the string on success, or -1 on failure</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern int llama_model_desc(SafeLlamaModelHandle model, byte* buf, long buf_size);
+        public static extern unsafe int llama_model_desc(SafeLlamaModelHandle model, byte* buf, long buf_size);
 
         /// <summary>
         /// Get the size of the model in bytes
@@ -493,7 +493,7 @@ namespace LLama.Native
         /// <param name="length">size of the buffer</param>
         /// <returns>The length written, or if the buffer is too small a negative that indicates the length required</returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern int llama_token_to_piece(SafeLlamaModelHandle model, int llamaToken, byte* buffer, int length);
+        public static extern unsafe int llama_token_to_piece(SafeLlamaModelHandle model, int llamaToken, byte* buffer, int length);
 
         /// <summary>
         /// Convert text into tokens
@@ -509,7 +509,7 @@ namespace LLama.Native
         /// Returns a negative number on failure - the number of tokens that would have been returned
         /// </returns>
         [DllImport(libraryName, CallingConvention = CallingConvention.Cdecl)]
-        public static extern int llama_tokenize(SafeLlamaModelHandle model, byte* text, int text_len, int* tokens, int n_max_tokens, bool add_bos, bool special);
+        public static extern unsafe int llama_tokenize(SafeLlamaModelHandle model, byte* text, int text_len, int* tokens, int n_max_tokens, bool add_bos, bool special);
 
         /// <summary>
         /// Register a callback to receive llama log messages

LLama/Native/NativeLibraryConfig.cs

@@ -29,10 +29,11 @@ namespace LLama.Native
         private bool _allowFallback = true;
         private bool _skipCheck = false;
         private bool _logging = false;
+
         /// <summary>
         /// search directory -> priority level, 0 is the lowest.
         /// </summary>
-        private List<string> _searchDirectories = new List<string>();
+        private readonly List<string> _searchDirectories = new List<string>();
 
         private static void ThrowIfLoaded()
         {
@@ -159,9 +160,8 @@ namespace LLama.Native
         internal static Description CheckAndGatherDescription()
         {
             if (Instance._allowFallback && Instance._skipCheck)
-            {
                 throw new ArgumentException("Cannot skip the check when fallback is allowed.");
-            }
+            
             return new Description(
                 Instance._libraryPath, 
                 Instance._useCuda, 
@@ -169,7 +169,8 @@ namespace LLama.Native
                 Instance._allowFallback, 
                 Instance._skipCheck, 
                 Instance._logging, 
-                Instance._searchDirectories.Concat(new string[] { "./" }).ToArray());
+                Instance._searchDirectories.Concat(new[] { "./" }).ToArray()
+            );
         }
 
         internal static string AvxLevelToString(AvxLevel level)
@@ -204,7 +205,9 @@ namespace LLama.Native
             if (!System.Runtime.Intrinsics.X86.X86Base.IsSupported)
                 return false;
 
+            // ReSharper disable UnusedVariable (ebx is used when < NET8)
             var (_, ebx, ecx, _) = System.Runtime.Intrinsics.X86.X86Base.CpuId(7, 0);
+            // ReSharper restore UnusedVariable
 
             var vnni = (ecx & 0b_1000_0000_0000) != 0;
 

LLama/Native/SafeLLamaContextHandle.cs

@@ -1,6 +1,5 @@
 using System;
 using System.Buffers;
-using System.Collections.Generic;
 using System.Text;
 using LLama.Exceptions;
 
@@ -51,8 +50,6 @@ namespace LLama.Native
             _model.DangerousAddRef(ref success);
             if (!success)
                 throw new RuntimeError("Failed to increment model refcount");
-
-            
         }
 
         /// <inheritdoc />

LLama/Native/SafeLlamaModelHandle.cs

@@ -214,7 +214,6 @@ namespace LLama.Native
         /// Get the metadata key for the given index
         /// </summary>
         /// <param name="index">The index to get</param>
-        /// <param name="buffer">A temporary buffer to store key characters in. Must be large enough to contain the key.</param>
         /// <returns>The key, null if there is no such key or if the buffer was too small</returns>
         public Memory<byte>? MetadataKeyByIndex(int index)
         {
@@ -243,7 +242,6 @@ namespace LLama.Native
         /// Get the metadata value for the given index
         /// </summary>
         /// <param name="index">The index to get</param>
-        /// <param name="buffer">A temporary buffer to store value characters in. Must be large enough to contain the value.</param>
         /// <returns>The value, null if there is no such value or if the buffer was too small</returns>
         public Memory<byte>? MetadataValueByIndex(int index)
         {