README

Azure OpenAI integration for Laravel. Chat, multi-turn conversations, streaming, multi-key rotation, model syncing, cost tracking, and powerful CLI tools — all built on ubxty/core-ai.

Requirements

PHP 8.2+
Laravel 11 or 12
An Azure OpenAI resource with at least one deployment

Installation

composer require ubxty/azure-ai

Publish the config file:

php artisan vendor:publish --tag=azure-ai-config

Run the migrations:

php artisan migrate

Configuration

Add the following to your .env:

AZURE_OPENAI_ENDPOINT=https://your-resource.openai.azure.com
AZURE_OPENAI_API_KEY=your-api-key
AZURE_OPENAI_API_VERSION=2024-10-21
AZURE_OPENAI_DEFAULT_MODEL=gpt-4o

For multi-key rotation, publish the config and define multiple keys under connections.default.keys.

Usage

Facade

use Ubxty\AzureAi\Facades\Azure;

// Single-turn invocation
$result = Azure::invoke(
    modelId: 'gpt-4o',
    systemPrompt: 'You are a helpful assistant.',
    userMessage: 'Explain recursion in simple terms.',
);

echo $result['response'];
echo $result['cost']; // in USD

Multi-turn conversation

use Ubxty\AzureAi\Facades\Azure;

$result = Azure::converse(
    modelId: 'gpt-4o',
    messages: [
        ['role' => 'user', 'content' => 'What is the capital of France?'],
        ['role' => 'assistant', 'content' => 'Paris.'],
        ['role' => 'user', 'content' => 'And Germany?'],
    ],
    systemPrompt: 'You are a geography expert.',
);

Streaming

Azure::stream(
    modelId: 'gpt-4o',
    messages: [['role' => 'user', 'content' => 'Tell me a story.']],
    onChunk: function (string $chunk) {
        echo $chunk;
        ob_flush();
    },
);

ConversationBuilder

use Ubxty\AzureAi\Facades\Azure;

Azure::conversation()
    ->model('gpt-4o')
    ->system('You are a helpful assistant.')
    ->user('What is PHP?')
    ->send();

Artisan Commands

Command	Description
`azure:chat`	Interactive multi-turn chat session in the terminal
`azure:configure`	Interactive wizard to write Azure credentials to `.env`
`azure:models`	List available deployments grouped by model family
`azure:test`	Run a test invocation and display response, tokens, and cost
`azure:default-model`	Set the default chat/image model in `.env`

Events

Event	Fired when
`AzureInvoked`	After every successful invocation
`AzureKeyRotated`	When a rate-limited key is rotated out
`AzureRateLimited`	When all keys are rate-limited

Health Check

Enable the built-in health check endpoint in your config:

'health_check' => [
    'enabled' => true,
    'path' => '/health/azure-openai',
    'middleware' => ['auth:sanctum'],
],

Changelog

See CHANGELOG.md.

License

MIT — see LICENSE.

ubxty / azure-ai

Maintainers

Package info

Statistics

Security