Using PHP with AI models hosted on GitHub

Overview

In this article I will show you how you can experiment with AI models hosted on GitHub in a simple PHP web app. GitHub AI Models are intended for learning, experimentation and proof-of-concept activities. The feature is subject to various limits (including requests per minute, requests per day, tokens per request, and concurrent requests) and is not designed for production use cases.

Prerequisites

To proceed, you will need the following:

• PHP - You need to have PHP version 8.3 (or higher) installed on your computer. You can download the latest version from https://www.php.net/downloads.php.
• Composer – If you do not have Composer yet, download and install it for your operating system from https://getcomposer.org/download/.

Getting Started

There are many AI models from a variety of vendors that you can choose from. The starting point is to visit https://github.com/marketplace/models. At the time of writing, these are a subset of the models available:

For this article, I will use the "DeepSeek-R1" beside the red arrow above. If you click on that model, you will be taken to the model's landing page:

Click on the green "Get API key" button.

The first thing we need to do is get a 'personal access token' by clicking on the “Get developer key” button.

Choose 'Generate new token', which happens to be in beta at the time of writing.

Give your token a name, set the expiration, and optionally describe the purpose of the token. Thereafter, click on the green 'Generate token' button at the bottom of the page.

Copy the newly generated token and place it is a safe place because you cannot view this token again once you leave the above page.

Let's do some PHP coding

In a working directory, create a sub-directory named PHP-GitHub-AI inside a terminal window with the following command:

mkdir PHP-GitHub-AI

Change into the newly created directory named PHP-GitHub-AI with:

cd PHP-GitHub-AI

In the PHP-GitHub-AI folder, create a file named index.php and add to it the following code:

<?php

// Set your Azure API key and endpoint
$apiKey = 'PUT-YOUR-PERSONAL-ACCESS-TOKEN-FROM-GITHUB-HERE';
$endpoint = 'https://models.inference.ai.azure.com';

// Define the API endpoint
$url = $endpoint . '/chat/completions';

// Set up the data for the API request

$data = [

    'messages' => [

        [

            'role' => 'system',

            'content' => 'you are an expert in astronomy'

        ],

        [

            'role' => 'user',

            'content' => 'which is the furthest planet to earth?'

        ]

    ],

    'model' => 'DeepSeek-R1',    // gpt-4o   DeepSeek-R1

    'temperature' => 1,

    'max_tokens' => 4096,

    'top_p' => 1

];

// Initialize cURL

$ch = curl_init($url);

// Set cURL options
curl_setopt($ch, CURLOPT_RETURNTRANSFER, true);
curl_setopt($ch, CURLOPT_HTTPHEADER, [
    'Content-Type: application/json',
    'Authorization: ' . $apiKey,
]);

curl_setopt($ch, CURLOPT_POST, true);
curl_setopt($ch, CURLOPT_POSTFIELDS, json_encode($data));

// Execute the API request
$response = curl_exec($ch);

// Check for errors

if ($response === false) {

    echo 'Error: ' . curl_error($ch);

} else {

    // Decode the response
    $result = json_decode($response, true);

    // Print the entire response for debugging

    /*

    echo '<pre>';

    print_r($result);

    echo '</pre>';

    */

    // Check if the 'choices' key exists in the response

    if (isset($result['choices'][0]['message']['content'])) {

        echo '<h3>Generated Text by ' . $result['model'] .':</h3>';

        // Print the generated text

        echo "<p>" . $result['choices'][0]['message']['content'] . "</p>";

    } else {

        if (isset($result['error'])) {

            echo 'Error: ' . $result['error']['message'];

        } else {

            echo 'Error: Unable to retrieve generated text.';

        }

    }

}

// Close cURL

curl_close($ch);

?>

In the above code:

• Set the value of $apiKey to be the personal access token from GitHub
• The system prompt is: 'you are an expert in astronomy'.
• The user prompt is: 'which is the furthest planet to earth?'
• We will be using the ‘DeepSeek-R1’ model

You can start the PHP web server in the PHP-GitHub-AI folder with this terminal window command:

php -S localhost:8888

Point your browser to http://localhost:8888. The output would look like this:

The output is in markdown format. We will need a library that converts from markdown to HTML. To that end, stop the web server and install the erusev/parsedown package into your application with this terminal window command:

composer require erusev/parsedown

Back in the code, make the following changes:

• Add this code to the first line of your PHP code:

// composer require erusev/parsedown

require_once 'vendor/autoload.php';

• Replace the following statement:

echo "<p>" . $result['choices'][0]['message']['content'] . "</p>";

WITH

error_reporting(E_ALL ^ E_DEPRECATED);

$Parsedown = new Parsedown();

$text =  $Parsedown->text($result['choices'][0]['message']['content']);

echo "<p>$text</p>";

Restart the web server with “php -S localhost:8888”. The page now shows a much better looking output:

You can change the AI model from DeepSeek-R1 to any other model (like gpt-4o) and will get similar results.