ML Ecosystem in PHP

Learning examples

These examples will help you understand how you can use ML in PHP. They are not full-fledged applications, but they will help you understand the basics of working with ML in PHP.

Example with Neuron AI

Neuron AI is another way to work with LLMs from PHP. The snippet below creates an agent, sends a user message, and prints the returned response content.

Example of use


                
<?php

namespace App\Neuron;

use NeuronAI\Agent\Agent;
use NeuronAI\Chat\Messages\UserMessage;
use NeuronAI\Providers\AIProviderInterface;
use NeuronAI\Providers\OpenAI\OpenAI;

/** @psalm-suppress PropertyNotSetInConstructor */
class MyAgent extends Agent {
    public function __construct() {
        parent::__construct();
    }

    protected function provider(): AIProviderInterface {
        return new OpenAI(
            key:  config('OPENAI_API_KEY', ''),
            model: config('OPENAI_API_MODEL', ''),
            parameters: [], // Add custom params (temperature, logprobs, etc)
            strict_response: false, // Strict structured output
        );
    }
}

$message = MyAgent::make()
    ->chat(new UserMessage('Hi!'))
    ->getMessage();


echo "Q1: Hi!\n";
echo 'A: ' . $message->getContent() . "\n\n";
echo "Q2: Explain who you are?\n";
echo 'A: ' . MyAgent::make()
    ->chat(new UserMessage('Explain who you are?'))
    ->getMessage()
    ->getContent();

Result: Memory: 0.001 Mb Time running: < 0.001 sec.


                Q1: Hi!
A: Hello! How can I assist you today?

Q2: Explain who you are?
A: Hello! I’m an AI assistant created using Neuron AI, which is a powerful agentic framework designed for the PHP ecosystem. My purpose is to help answer your questions, provide explanations, assist with coding tasks, and support you in various topics, especially related to PHP development. How can I assist you today?