Skip to main content

​
OpenAI Provider

The OpenAI provider gives you access to OpenAI’s latest language models, including the GPT-4o series and advanced reasoning models like o1 and o3.

​
Configuration

​
API Key Setup

Set your OpenAI API key in your .env file: 
OPENAI_API_KEY=your_api_key_here
OPENAI_ORGANIZATION_ID=your_org_id_here  # Optional
The API key configuration is defined in config/llm-magic.php: 
'apis' => [
    'openai' => [
        'token' => env('OPENAI_API_KEY'),
        'organization_id' => env('OPENAI_ORGANIZATION_ID'),
    ],
]

​
Available Models

LLM Magic supports the following OpenAI models:

​
Reasoning Models

  • o1 - Advanced reasoning model
  • o1-pro - Professional tier reasoning model
  • o3-mini - Compact o3 model
  • o3 - Latest reasoning model
  • o4-mini - Compact o4 model

​
GPT-4 Series

  • gpt-4-turbo - GPT-4 Turbo
  • gpt-4o - GPT-4 Optimized
  • gpt-4o-mini - Compact GPT-4o model

​
Usage

​
Using the Constructor

Create an OpenAI model instance with any model name: 
use Mateffy\Magic\Models\OpenAI;
use Mateffy\Magic\Models\Options\ChatGptOptions;

$model = new OpenAI(
    model: 'gpt-4o',
    options: new ChatGptOptions
);

​
Using Static Factory Methods

LLM Magic provides convenient static methods for common models: 
use Mateffy\Magic\Models\OpenAI;

// GPT-4o Mini - fastest and most cost-effective
$model = OpenAI::gpt_4o_mini();

// GPT-4o - balanced performance
$model = OpenAI::gpt_4o();

// GPT-4 Turbo - high performance
$model = OpenAI::gpt_4_turbo();

// o1 models - advanced reasoning
$model = OpenAI::o1_mini();
$model = OpenAI::o1_preview();

​
Getting Available Models

Retrieve a list of all available OpenAI models: 
use Mateffy\Magic\Models\OpenAI;

$models = OpenAI::models();
// Returns a Collection with prefixed model names like 'openai/gpt-4o'

$models = OpenAI::models(prefix: null, prefixLabels: null);
// Returns models without prefix

​
Model Costs

OpenAI models include built-in cost tracking. The following pricing is configured (per 1,000 tokens):
ModelInput CostOutput Cost
o1$0.015$0.075
o1-pro$0.003$0.015
o3-mini$0.0008$0.004
o3$0.0008$0.004
o4-mini$0.0008$0.004
gpt-4-turbo$0.0008$0.004
gpt-4o$0.0008$0.004
gpt-4o-mini$0.0008$0.004
Access cost information programmatically: 
$model = OpenAI::gpt_4o();
$cost = $model->getModelCost();

if ($cost) {
    echo "Input: {$cost->inputCentsPer1K} cents per 1K tokens";
    echo "Output: {$cost->outputCentsPer1K} cents per 1K tokens";
}

​
Organization Info

The OpenAI provider includes organization metadata:
  • ID: openai
  • Name: OpenAI
  • Website: https://openai.com
  • Privacy: Data may be used for model training and abuse prevention

​
Advanced Options

You can pass additional options using ChatGptOptions: 
use Mateffy\Magic\Models\OpenAI;
use Mateffy\Magic\Models\Options\ChatGptOptions;

$options = new ChatGptOptions(
    // Configure temperature, max tokens, etc.
);

$model = new OpenAI('gpt-4o', $options);

​
See Also

Build docs developers (and LLMs) love

Get started for freeTalk to us