rodiumai/laravel-sdk

Official PHP / Laravel SDK for the Rodium AI API — unified access to AI models (OpenAI, Anthropic, Google, DeepSeek, MiniMax…) with RODI credit billing and Mobile Money top-ups.

OpenAI-compatible REST API: same endpoints and payloads as documented at rodiumai.io/docs.

Links

Table of contents

• Official documentation
• Requirements
• Installation
• Configuration
• Quick start
• Dynamic models (API)
• Multilingual
• Chat parameters
• Streaming (SSE)
• Listing models
• Error handling
• SDK reference
• Testing & development
• Contributing
• License

Official documentation

Detailed SDK ↔ API mapping: docs/api-alignment.md.

Requirements

• PHP 8.1+ with the json extension
• Laravel 10, 11, 12, or 13 (optional — the client works in plain PHP)
• Laravel 12+: PHP 8.2+ → use ^0.1.1 minimum
• Rodium AI account + API key: dashboard

Installation

Install from Packagist:

composer require rodiumai/laravel-sdk

For Laravel 12 and 13 (PHP 8.2+):

composer require rodiumai/laravel-sdk:^0.1.1

After installation

1. Publish config (optional but recommended):

php artisan vendor:publish --tag=rodiumai-config

1. Add your API key to .env (see Configuration).

2. The ServiceProvider and RodiumAI Facade are auto-discovered — nothing to register in bootstrap/providers.php.

Plain PHP (no Laravel)

Same Composer command. Then instantiate RodiumAI\RodiumAIClient directly (see Quick start).

Configuration

.env file:

RODIUMAI_API_KEY=rd_sk_your_secret_key
RODIUMAI_DEFAULT_MODEL=openai/gpt-4o
RODIUMAI_TIMEOUT=30
# RODIUMAI_LOCALE=fr   # optional — SDK error hints only

The SDK always uses https://api.rodiumai.io/v1 (not configurable).

Never commit .env or API keys to the repository.

Quick start

Laravel (Facade)

use RodiumAI\Facades\RodiumAI;

$catalogue = RodiumAI::models();
$modelId = $catalogue->chatModels()[0]->id;

$response = RodiumAI::model($modelId)
    ->temperature(0.7)
    ->maxTokens(300)
    ->chat('Explain Rodium AI in two sentences.');

echo $response->content;

Plain PHP

use RodiumAI\RodiumAIClient;

$client = new RodiumAIClient(apiKey: getenv('RODIUMAI_API_KEY'));

$catalogue = $client->models();
$modelId = $catalogue->chatModels()[0]->id;

$response = $client->model($modelId)->chat('Hello!');

echo $response->content;

Equivalent to the cURL / OpenAI SDK quickstart: POST https://api.rodiumai.io/v1/chat/completions with Authorization: Bearer {RODIUMAI_API_KEY}.

Dynamic models (API)

Models are fetched live from GET /v1/models — always up to date with the platform catalogue:

$catalogue = RodiumAI::models();

$catalogue->ids();                    // all model IDs
$catalogue->chatModels();             // text/chat-capable only
$catalogue->providerPrefixes();       // ['anthropic', 'google', 'openai', …]
$catalogue->byProvider('anthropic');  // filter by provider prefix
$catalogue->findById('openai/gpt-4o'); // ?ModelInfo

foreach ($catalogue->chatModels() as $info) {
    echo $info->id . ' — ' . $info->contextWindow . PHP_EOL;
}

Pass any id from the API to ->model('openai/gpt-4o').

Multilingual (optional)

Language is never required. By default, SDK error hints are in English and the AI follows the user's message language.

Optional — localized error hints via config:

RODIUMAI_LOCALE=fr

Optional — force AI responses in a specific language:

RodiumAI::language('fr')  // only when you need it
    ->model('openai/gpt-4o')
    ->chat('Bonjour !');

Supported: en (default), fr, es.

Chat parameters

Aligned with chat-completions:

use RodiumAI\Data\ChatMessage;

$messages = [
    ChatMessage::system('You are a Laravel assistant.'),
    ChatMessage::user('What is a Service Provider?'),
];

$response = RodiumAI::model('openai/gpt-4o')
    ->temperature(0.5)
    ->topP(0.9)
    ->maxTokens(500)
    ->chat($messages);

Streaming (SSE)

Follows docs/api/streaming: data: … lines, end with data: [DONE].

foreach (RodiumAI::model('openai/gpt-4o')->stream('Tell a short story.') as $delta) {
    echo $delta;
}

Laravel — StreamedResponse

use Illuminate\Http\Request;
use Symfony\Component\HttpFoundation\StreamedResponse;
use RodiumAI\Facades\RodiumAI;

public function streamChat(Request $request): StreamedResponse
{
    return response()->stream(function () use ($request) {
        foreach (RodiumAI::stream($request->string('message')) as $delta) {
            echo 'data: ' . json_encode(['delta' => $delta]) . "\n\n";
            ob_flush();
            flush();
        }
        echo "data: [DONE]\n\n";
    }, 200, [
        'Content-Type' => 'text/event-stream',
        'Cache-Control' => 'no-cache',
        'X-Accel-Buffering' => 'no',
    ]);
}

Listing models

GET /v1/models — RODI pricing and metadata:

$models = RodiumAI::models();

foreach ($models->ids() as $id) {
    echo $id . PHP_EOL;
}

$anthropic = $models->byProvider('anthropic');

Error handling

See docs/api/errors.

use RodiumAI\Exceptions\InsufficientCreditsException;
use RodiumAI\Exceptions\RodiumAIException;
use RodiumAI\Facades\RodiumAI;

try {
    $response = RodiumAI::chat('Test');
} catch (InsufficientCreditsException $e) {
    logger()->warning('Insufficient RODI', ['body' => $e->responseBody()]);
} catch (RodiumAIException $e) {
    logger()->error('Rodium AI', ['code' => $e->getCode(), 'body' => $e->responseBody()]);
}

SDK reference

DTOs: ChatResponse, ChatMessage, ModelCollection.

Versions

Full history: CHANGELOG.md.

Testing & development

composer install
composer test                 # PHPUnit (mocked HTTP)
export RODIUMAI_API_KEY="…"
php bin/smoke-test.php        # Live API walkthrough in the terminal

Contributing

See CONTRIBUTING.md and docs/architecture.md.

Maintainers: docs/PUBLISHING.md (tags, Packagist).

License

MIT — see LICENSE.