Darkspyre.ML.Tokenizers 2.0.1.1

Prefix Reserved
dotnet add package Darkspyre.ML.Tokenizers --version 2.0.1.1
                    
NuGet\Install-Package Darkspyre.ML.Tokenizers -Version 2.0.1.1
                    
This command is intended to be used within the Package Manager Console in Visual Studio, as it uses the NuGet module's version of Install-Package.
<PackageReference Include="Darkspyre.ML.Tokenizers" Version="2.0.1.1" />
                    
For projects that support PackageReference, copy this XML node into the project file to reference the package.
<PackageVersion Include="Darkspyre.ML.Tokenizers" Version="2.0.1.1" />
                    
Directory.Packages.props
<PackageReference Include="Darkspyre.ML.Tokenizers" />
                    
Project file
For projects that support Central Package Management (CPM), copy this XML node into the solution Directory.Packages.props file to version the package.
paket add Darkspyre.ML.Tokenizers --version 2.0.1.1
                    
#r "nuget: Darkspyre.ML.Tokenizers, 2.0.1.1"
                    
#r directive can be used in F# Interactive and Polyglot Notebooks. Copy this into the interactive tool or source code of the script to reference the package.
#:package Darkspyre.ML.Tokenizers@2.0.1.1
                    
#:package directive can be used in C# file-based apps starting in .NET 10 preview 4. Copy this into a .cs file before any lines of code to reference the package.
#addin nuget:?package=Darkspyre.ML.Tokenizers&version=2.0.1.1
                    
Install as a Cake Addin
#tool nuget:?package=Darkspyre.ML.Tokenizers&version=2.0.1.1
                    
Install as a Cake Tool

Darkspyre fork

Darkspyre.ML.Tokenizers is a source-compatible fork of Microsoft's tokenizer library. Version 2.0.1.1 is based on Microsoft.ML.Tokenizers 2.0.1 and corrects SentencePiece models that disable BOS or EOS with token id -1. Assemblies, namespaces, and public APIs retain their upstream names.

About

Microsoft.ML.Tokenizers provides an abstraction for tokenizers as well as implementations of common tokenization algorithms.

Key Features

  • Extensible tokenizer architecture that allows for specialization of Normalizer, PreTokenizer, Model/Encoder, Decoder
  • BPE - Byte pair encoding model
  • English Roberta model
  • Tiktoken model
  • Llama model
  • Phi2 model

How to Use

using Microsoft.ML.Tokenizers;
using System.IO;
using System.Net.Http;

//
// Using Tiktoken Tokenizer
//

// Initialize the tokenizer for the `gpt-4o` model. This instance should be cached for all subsequent use.
Tokenizer tokenizer = TiktokenTokenizer.CreateForModel("gpt-4o");

string source = "Text tokenization is the process of splitting a string into a list of tokens.";

Console.WriteLine($"Tokens: {tokenizer.CountTokens(source)}");
// prints: Tokens: 16

var trimIndex = tokenizer.GetIndexByTokenCountFromEnd(source, 5, out string processedText, out _);
Console.WriteLine($"5 tokens from end: {processedText.Substring(trimIndex)}");
// prints: 5 tokens from end:  a list of tokens.

trimIndex = tokenizer.GetIndexByTokenCount(source, 5, out processedText, out _);
Console.WriteLine($"5 tokens from start: {processedText.Substring(0, trimIndex)}");
// prints: 5 tokens from start: Text tokenization is the

IReadOnlyList<int> ids = tokenizer.EncodeToIds(source);
Console.WriteLine(string.Join(", ", ids));
// prints: 1199, 4037, 2065, 374, 279, 1920, 315, 45473, 264, 925, 1139, 264, 1160, 315, 11460, 13

//
// Using Llama Tokenizer
//

// Open a stream to the remote Llama tokenizer model data file.
using HttpClient httpClient = new();
const string modelUrl = @"https://huggingface.co/hf-internal-testing/llama-tokenizer/resolve/main/tokenizer.model";
using Stream remoteStream = await httpClient.GetStreamAsync(modelUrl);

// Create the Llama tokenizer using the remote stream. This should be cached for all subsequent use.
Tokenizer llamaTokenizer = LlamaTokenizer.Create(remoteStream);

string input = "Hello, world!";
ids = llamaTokenizer.EncodeToIds(input);
Console.WriteLine(string.Join(", ", ids));
// prints: 1, 15043, 29892, 3186, 29991

Console.WriteLine($"Tokens: {llamaTokenizer.CountTokens(input)}");
// prints: Tokens: 5

Main Types

The main types provided by this library are:

  • Microsoft.ML.Tokenizers.Tokenizer
  • Microsoft.ML.Tokenizers.BpeTokenizer
  • Microsoft.ML.Tokenizers.EnglishRobertaTokenizer
  • Microsoft.ML.Tokenizers.TiktokenTokenizer
  • Microsoft.ML.Tokenizers.Normalizer
  • Microsoft.ML.Tokenizers.PreTokenizer

Additional Documentation

Feedback & Contributing

Microsoft.ML.Tokenizers is released as open source under the MIT license. Bug reports and contributions are welcome at the GitHub repository.

Product Compatible and additional computed target framework versions.
.NET net5.0 was computed.  net5.0-windows was computed.  net6.0 was computed.  net6.0-android was computed.  net6.0-ios was computed.  net6.0-maccatalyst was computed.  net6.0-macos was computed.  net6.0-tvos was computed.  net6.0-windows was computed.  net7.0 was computed.  net7.0-android was computed.  net7.0-ios was computed.  net7.0-maccatalyst was computed.  net7.0-macos was computed.  net7.0-tvos was computed.  net7.0-windows was computed.  net8.0 is compatible.  net8.0-android was computed.  net8.0-browser was computed.  net8.0-ios was computed.  net8.0-maccatalyst was computed.  net8.0-macos was computed.  net8.0-tvos was computed.  net8.0-windows was computed.  net9.0 was computed.  net9.0-android was computed.  net9.0-browser was computed.  net9.0-ios was computed.  net9.0-maccatalyst was computed.  net9.0-macos was computed.  net9.0-tvos was computed.  net9.0-windows was computed.  net10.0 was computed.  net10.0-android was computed.  net10.0-browser was computed.  net10.0-ios was computed.  net10.0-maccatalyst was computed.  net10.0-macos was computed.  net10.0-tvos was computed.  net10.0-windows was computed. 
.NET Core netcoreapp2.0 was computed.  netcoreapp2.1 was computed.  netcoreapp2.2 was computed.  netcoreapp3.0 was computed.  netcoreapp3.1 was computed. 
.NET Standard netstandard2.0 is compatible.  netstandard2.1 was computed. 
.NET Framework net461 was computed.  net462 was computed.  net463 was computed.  net47 was computed.  net471 was computed.  net472 was computed.  net48 was computed.  net481 was computed. 
MonoAndroid monoandroid was computed. 
MonoMac monomac was computed. 
MonoTouch monotouch was computed. 
Tizen tizen40 was computed.  tizen60 was computed. 
Xamarin.iOS xamarinios was computed. 
Xamarin.Mac xamarinmac was computed. 
Xamarin.TVOS xamarintvos was computed. 
Xamarin.WatchOS xamarinwatchos was computed. 
Compatible target framework(s)
Included target framework(s) (in package)
Learn more about Target Frameworks and .NET Standard.

NuGet packages (11)

Showing the top 5 NuGet packages that depend on Darkspyre.ML.Tokenizers:

Package Downloads
Darkspyre.Inference.Tensor.Registry

Registration, validation and ordered discovery of the Darkspyre Tensor engine implementations.

Darkspyre.Inference.Connect

Image, speech and audio inference connectors for Darkspyre applications.

Darkspyre.Inference.Tensor.Decision.Laya

Laya in-process typed-decision inference through ONNX Runtime.

Darkspyre.Inference.Tensor.Image.Supra

Supra2-IMG in-process image generation through TensorSharp.

Darkspyre.Inference.Connect.Common

Universal conversation and embedding contracts, shared provider infrastructure and automatic Connect discovery.

GitHub repositories

This package is not used by any popular GitHub repositories.

Version Downloads Last Updated
2.0.1.1 346 10/18/2026

Based on Microsoft.ML.Tokenizers 2.0.1; accepts SentencePiece models whose BOS or EOS token is disabled with id -1.