Appearance

XAI (Grok) Provider ​

Overview ​

XAI is the AI company behind Grok, a family of large language models known for their conversational abilities and real-time knowledge. Since XAI's API is OpenAI-compatible, Mindwave reuses its OpenAI driver with a custom base URI, giving you access to all OpenAI driver features with Grok models.

Why Use XAI with Mindwave? ​

  • Strong conversational models - Grok excels at natural, engaging conversations
  • OpenAI-compatible API - Drop-in replacement using the same interface
  • Competitive pricing - Cost-effective alternative for chat applications
  • Streaming support - Full SSE streaming through the OpenAI driver
  • All OpenAI features - Function calling, JSON mode, and more

Key Capabilities ​

  • ✅ Chat completions with multi-turn conversations
  • ✅ Streaming responses with SSE
  • ✅ Function/tool calling
  • ✅ JSON mode for structured outputs
  • ✅ Automatic tracing and cost tracking
  • ✅ All OpenAI driver features

Setup & Configuration ​

Getting Your API Key ​

  1. Visit console.x.ai
  2. Create an account or sign in
  3. Navigate to API Keys
  4. Generate a new API key
  5. Copy your key

Environment Variables ​

Add these to your .env file:

dotenv
# Required: Your XAI API key
XAI_API_KEY=your-xai-api-key-here

# Optional: Default model
MINDWAVE_XAI_MODEL=grok-3-mini

# Optional: Generation parameters
MINDWAVE_XAI_MAX_TOKENS=1000
MINDWAVE_XAI_TEMPERATURE=0.4

# Set XAI as default provider
MINDWAVE_LLM=xai

Configuration File ​

The XAI configuration is defined in config/mindwave-llm.php:

php
return [
    'default' => env('MINDWAVE_LLM', 'openai'),

    'llms' => [
        'xai' => [
            'api_key' => env('XAI_API_KEY'),
            'model' => env('MINDWAVE_XAI_MODEL', 'grok-3-mini'),
            'max_tokens' => env('MINDWAVE_XAI_MAX_TOKENS', 1000),
            'temperature' => env('MINDWAVE_XAI_TEMPERATURE', 0.4),
        ],
    ],
];

How It Works ​

XAI uses the OpenAI-compatible API format, so Mindwave creates an OpenAI driver instance with XAI's base URI:

php
// Internally, Mindwave does this:
$client = OpenAI::factory()
    ->withApiKey(config('mindwave-llm.llms.xai.api_key'))
    ->withBaseUri('https://api.x.ai/v1')
    ->make();

This means all OpenAI driver features work seamlessly with XAI.

Testing Your Connection ​

php
use Mindwave\Mindwave\Facades\LLM;

$response = LLM::driver('xai')->generateText('Say hello!');

if ($response) {
    echo "XAI connection successful!";
    echo "Response: " . $response;
} else {
    echo "Connection failed - check your API key";
}

Available Models ​

A fast, cost-effective model suitable for most tasks.

Model IDContext WindowBest For
grok-3-mini131,072 tokensGeneral chat, fast responses

Grok 3 ​

The full-size Grok model for complex tasks.

Model IDContext WindowBest For
grok-3131,072 tokensComplex reasoning, detailed outputs

Check Available Models

XAI regularly updates their model lineup. Check docs.x.ai for the latest available models and pricing.

Basic Usage ​

Simple Text Generation ​

php
use Mindwave\Mindwave\Facades\LLM;

$response = LLM::driver('xai')
    ->generateText('Explain Laravel in one sentence.');

echo $response;

Chat Completion ​

php
use Mindwave\Mindwave\Facades\LLM;

$response = LLM::driver('xai')
    ->model('grok-3-mini')
    ->maxTokens(500)
    ->temperature(0.7)
    ->chat([
        ['role' => 'system', 'content' => 'You are a helpful Laravel expert.'],
        ['role' => 'user', 'content' => 'How do I create a migration?'],
    ]);

echo $response->content;

Streaming Responses ​

Since XAI uses the OpenAI driver, streaming works identically to OpenAI:

php
use Mindwave\Mindwave\Facades\LLM;
use Mindwave\Mindwave\LLM\Streaming\StreamedTextResponse;

Route::get('/api/chat', function (Request $request) {
    $stream = LLM::driver('xai')
        ->model('grok-3-mini')
        ->streamText($request->input('prompt'));

    $response = new StreamedTextResponse($stream);

    return $response->toStreamedResponse();
});

Text Streaming ​

php
foreach (LLM::driver('xai')->streamText('Write a story') as $chunk) {
    echo $chunk;
}

Best Practices ​

When to Use XAI ​

Choose XAI when:

  • You want Grok's conversational style
  • You need a solid alternative to OpenAI
  • Cost-effective chat applications

Consider other providers when:

  • You need the absolute best reasoning (use OpenAI GPT-4o or Anthropic Claude)
  • You need very large context windows (Gemini 1.5 Pro offers 2M tokens)
  • You need function calling with complex tools (OpenAI has the most mature support)

Troubleshooting ​

401 Unauthorized ​

Solutions:

  1. Verify your API key at console.x.ai
  2. Check .env:
    dotenv
    XAI_API_KEY=your-actual-key
  3. Clear config cache: php artisan config:clear

Model Not Available ​

Some Grok models may require specific account tiers. Check your account access at console.x.ai.

Summary ​

XAI provides access to Grok models through an OpenAI-compatible API:

  • Grok 3 Mini - Fast and cost-effective for most tasks
  • Grok 3 - Full-size model for complex reasoning
  • OpenAI-compatible - Same API, same code, different model

Next Steps:

Resources: