Add Microsoft.ML.Tokenizers documentation to .NET AI docs

[Copilot]

Adds comprehensive how-to guide for the Microsoft.ML.Tokenizers library, addressing the gap in documentation for this ML.NET tokenization package that's increasingly used in AI scenarios.

Changes

• New how-to article at docs/ai/how-to/use-tokenizers.md

  • Installation instructions including required data packages
  • Tiktoken tokenizer examples (GPT-4/OpenAI models)
  • Llama tokenizer examples with Hugging Face model loading
  • BPE tokenization demonstrations
  • Advanced encoding options (normalization, pretokenization control)
  • Migration guidance from DeepDev.TokenizerLib and SharpToken

• Working code examples in docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/

  • Complete, compilable project with all tokenizer types
  • Demonstrates token counting, encoding/decoding, and text trimming operations

• TOC and cross-references

  • Added "Tokenization" section to AI docs TOC
  • Linked from existing "Understanding tokens" conceptual article

Example usage

// Initialize tokenizer for GPT-4o (cache this instance) Tokenizer tokenizer = TiktokenTokenizer.CreateForModel("gpt-4o"); string text = "Text tokenization is the process of splitting a string into a list of tokens."; // Count tokens for cost estimation Console.WriteLine($"Tokens: {tokenizer.CountTokens(text)}"); // Output: 16 // Encode to token IDs IReadOnlyList<int> ids = tokenizer.EncodeToIds(text); // Trim text to token limit var index = tokenizer.GetIndexByTokenCount(text, 5, out string? processed, out _); Console.WriteLine($"First 5 tokens: {processed.Substring(0, index)}");

Warning

Firewall rules blocked me from connecting to one or more addresses (expand for details)

I tried to connect to the following addresses, but was blocked by firewall rules:

• huggingface.co
  • Triggering command: /home/REDACTED/work/docs/docs/docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/bin/Debug/net9.0/TokenizersExamples (dns block)

If you need me to access, download, or install something from one of these locations, you can either:

• Configure Actions setup steps to set up my environment, which run before the firewall is enabled
• Add the appropriate URLs or hosts to the custom allowlist in this repository's Copilot coding agent settings (admins only)

Original prompt

---

This section details on the original issue you should resolve

<issue_title>Add content for Microsoft.ML.Tokenizers</issue_title>
<issue_description>Add a conceptual doc for using the Microsoft.ML.Tokenizers package, which is technically part of the ML.NET set of libraries. That's an area we don't have a lot of documentation around besides what's in the NuGet README but has received more investment in the past few months.

The new article should either live with the ML.NET docs or the .NET AI docs.

Content from NuGet readme:

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

• Conceptual documentation
• API documentation</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes Add content for Microsoft.ML.Tokenizers #48455

---

💬 We'd love your input! Share your thoughts on Copilot coding agent in our 2 minute survey.

---

Internal previews

📄 File	🔗 Preview link
docs/ai/conceptual/understanding-tokens.md	Understand tokens
docs/ai/how-to/use-tokenizers.md	Use Microsoft.ML.Tokenizers for text tokenization
docs/ai/toc.yml	docs/ai/toc

[Copilot]

Pull Request Overview

This PR adds comprehensive documentation for the Microsoft.ML.Tokenizers library, introducing a new how-to guide that covers tokenization concepts and provides practical examples for working with different tokenizer models in .NET applications.

Key changes:

• New how-to article explaining Microsoft.ML.Tokenizers usage with examples for Tiktoken, Llama, and BPE tokenizers.
• Code examples demonstrating common tokenization operations including encoding, decoding, token counting, and text trimming.
• Integration with existing documentation through TOC updates and cross-references.

Reviewed Changes

Copilot reviewed 9 out of 9 changed files in this pull request and generated 7 comments.

Show a summary per file

File	Description
docs/ai/toc.yml	Adds new "Tokenization" section with link to the tokenizers guide
docs/ai/how-to/use-tokenizers.md	New how-to article covering Microsoft.ML.Tokenizers library usage
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/TokenizersExamples.csproj	Project file for tokenizer code examples
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/TiktokenExample.cs	Example code demonstrating Tiktoken tokenizer usage
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/Program.cs	Entry point for running all tokenizer examples
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/LlamaExample.cs	Example code for Llama tokenizer usage
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/BpeExample.cs	Example code for BPE tokenizer usage
docs/ai/how-to/snippets/use-tokenizers/csharp/TokenizersExamples/.gitignore	Ignores build artifacts for example project
docs/ai/conceptual/understanding-tokens.md	Adds cross-reference to new tokenizers guide