• Home
  • History
  • Annotate
Name Date Size #Lines LOC

..--

data/02-Apr-2025-50,00450,003

src/02-Apr-2025-288214

tests/02-Apr-2025-2923

.editorconfigD01-Apr-2025337 1914

.php-cs-fixer.dist.phpD01-Apr-2025212 1411

CHANGELOG.mdD01-Apr-2025327 1610

LICENSED01-Apr-20251 KiB2217

README.mdD01-Apr-2025625 2617

composer.jsonD01-Apr-20251.1 KiB4746

composer.lockD01-Apr-2025136.5 KiB3,9243,923

phpstan.neonD01-Apr-202566 65

rector.phpD01-Apr-2025569 2518

README.md

1# TikToken clone for PHP - PHP GPT3 tokenizer
2
3PHP Text Tokenizer for GPT models
4
5## About
6
7A PHP toolkit to tokenize text like GPT family of models process it.
8
9Forked from [semji/gpt3-tokenizer-php](https://github.com/semji/gpt3-tokenizer-php) to bug fixes and improvement.
10
11## Requirements
12* PHP 8.1
13* mbstring extension [details here on how to install mbstring](https://www.php.net/manual/en/mbstring.installation.php)
14## Usage
15
16```php
17use TikToken\Encoder;
18$prompt = "Ai is cool";
19$encoder = new Encoder();
20
21$tokens = $encoder->encode($prompt); // [32, 72, 318, 3608]
22
23// Get tokens count:
24echo count($tokens); // 4
25```
26