### Install PerplexityAI SDK via Composer

Source: https://github.com/softcreatr/php-perplexity-ai-sdk/blob/main/README.md

Installs the PerplexityAI PHP SDK using Composer, a dependency manager for PHP. This is the standard method for adding the library to your project.

```bash
composer require softcreatr/php-perplexity-ai-sdk
```

--------------------------------

### Initialize PHP Perplexity AI SDK

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Demonstrates how to initialize the PerplexityAI class in PHP. It requires setting up PSR-17 and PSR-18 compliant factories and an HTTP client, then instantiating the SDK with your API key.

```php
<?php

require_once 'vendor/autoload.php';

use SoftCreatR\PerplexityAI\PerplexityAI;
use GuzzleHttp\Client;
use GuzzleHttp\Psr7\HttpFactory;

// Initialize PSR-17 and PSR-18 implementations
$psr17Factory = new HttpFactory();
$httpClient = new Client(['stream' => true]);

// Create PerplexityAI instance
$pplx = new PerplexityAI(
    requestFactory: $psr17Factory,
    streamFactory: $psr17Factory,
    uriFactory: $psr17Factory,
    httpClient: $httpClient,
    apiKey: 'your_api_key_here'
);

```

--------------------------------

### Initialize PerplexityAI SDK in PHP

Source: https://github.com/softcreatr/php-perplexity-ai-sdk/blob/main/README.md

Demonstrates how to initialize the PerplexityAI SDK in PHP. It requires including the autoloader, setting up an API key, and instantiating the PerplexityAI class with necessary HTTP client and factory implementations.

```php
<?php

require_once 'vendor/autoload.php';

use SoftCreatR\PerplexityAI\PerplexityAI;

$apiKey = 'your_api_key';

// Replace these lines with your chosen PSR-17 and PSR-18 compatible HTTP client and factories
$httpClient = new YourChosenHttpClient();
$requestFactory = new YourChosenRequestFactory();
$streamFactory = new YourChosenStreamFactory();
$uriFactory = new YourChosenUriFactory();

$pplx = new PerplexityAI($requestFactory, $streamFactory, $uriFactory, $httpClient, $apiKey);

```

--------------------------------

### Configure Perplexity AI SDK with Custom Headers and Settings (PHP)

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Demonstrates how to initialize the PerplexityAI SDK with custom origin and base path, and send chat completion requests with custom HTTP headers. This allows for advanced customization of API interactions beyond default settings.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAI;

// Initialize with custom origin and base path
$pplx = new PerplexityAI(
    requestFactory: $psr17Factory,
    streamFactory: $psr17Factory,
    uriFactory: $psr17Factory,
    httpClient: $httpClient,
    apiKey: 'your_api_key',
    origin: 'custom.api.endpoint.com',  // Optional custom domain
    basePath: '/v1'                      // Optional custom base path
);

// Send request with custom headers
$response = $pplx->createChatCompletion([], [
    'model' => 'sonar',
    'messages' => [
        [
            'role' => 'user',
            'content' => 'What is quantum computing?'
        ]
    ],
    'customHeaders' => [
        'X-Custom-Header' => 'custom-value',
        'X-Request-ID' => 'req-12345'
    ]
]);

$result = json_decode($response->getBody()->getContents(), true);
echo $result['choices'][0]['message']['content'];

```

--------------------------------

### Build Perplexity AI URLs and Manage Endpoints (PHP)

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Illustrates how to use the PerplexityAIURLBuilder to retrieve endpoint configurations (method and path) and construct complete API URLs. This is useful for understanding API structure and for manual URL generation if needed.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAIURLBuilder;
use GuzzleHttp\Psr7\HttpFactory;

$uriFactory = new HttpFactory();

// Get endpoint configuration
$endpoint = PerplexityAIURLBuilder::getEndpoint('createChatCompletion');
echo "Method: " . $endpoint['method'] . "\n";  // Output: POST
echo "Path: " . $endpoint['path'] . "\n";      // Output: /chat/completions

// Build complete URL
$uri = PerplexityAIURLBuilder::createUrl(
    uriFactory: $uriFactory,
    key: 'createChatCompletion',
    parameters: [],
    origin: 'api.perplexity.ai',
    basePath: ''
);

echo "Full URL: " . (string)$uri . "\n";
// Output: https://api.perplexity.ai/chat/completions

```

--------------------------------

### Factory Pattern for Perplexity AI Client in PHP

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Implements a factory pattern to abstract the creation of the Perplexity AI client. This class centralizes the instantiation of the `PerplexityAI` object, including its HTTP client and PSR factories, and provides convenient static methods for making requests, handling both synchronous and streaming responses. It simplifies the usage of the SDK in larger applications by managing dependencies and providing a consistent interface for API calls.

```php
<?php

use GuzzleHttp\Client;
use GuzzleHttp\Psr7\HttpFactory;
use SoftCreatR\PerplexityAI\PerplexityAI;

class PerplexityAIFactory
{
    public static function create(string $apiKey): PerplexityAI
    {
        $psr17Factory = new HttpFactory();
        $httpClient = new Client(['stream' => true]);

        return new PerplexityAI(
            requestFactory: $psr17Factory,
            streamFactory: $psr17Factory,
            uriFactory: $psr17Factory,
            httpClient: $httpClient,
            apiKey: $apiKey
        );
    }

    public static function request(
        string $method,
        array $parameters = [],
        array $options = [],
        ?callable $streamCallback = null
    ): mixed {
        $pplx = self::create($_ENV['PERPLEXITY_API_KEY']);

        if ($streamCallback !== null) {
            $pplx->{$method}($parameters, $options, $streamCallback);
            return null;
        }

        $response = $pplx->{$method}($parameters, $options);
        return json_decode($response->getBody()->getContents(), true);
    }
}

// Usage
$result = PerplexityAIFactory::request(
    'createChatCompletion',
    [],
    [
        'model' => 'sonar',
        'messages' => [
            ['role' => 'user', 'content' => 'What is machine learning?']
        ]
    ]
);

echo $result['choices'][0]['message']['content'];

```

--------------------------------

### Use Advanced Parameters for Chat Completions in PHP

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Demonstrates how to utilize advanced parameters such as model selection, token limits, temperature, sampling, penalties, stop sequences, and citation/image return options when creating chat completions with the Perplexity AI SDK. This allows for fine-tuning the AI's response generation for specific use cases.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAI;

// Use advanced model parameters
$response = $pplx->createChatCompletion([], [
    'model' => 'llama-3.1-sonar-large-128k-online',
    'messages' => [
        [
            'role' => 'user',
            'content' => 'Explain quantum entanglement.'
        ]
    ],
    'max_tokens' => 500,              // Maximum tokens to generate
    'temperature' => 0.2,             // Lower = more focused, higher = more creative
    'top_p' => 0.9,                   // Nucleus sampling parameter
    'presence_penalty' => 0.0,        // Penalize new topics (-2.0 to 2.0)
    'frequency_penalty' => 0.0,       // Penalize repetition (-2.0 to 2.0)
    'stop' => ["\n\n", "###"],        // Stop sequences
    'return_citations' => true,        // Include source citations
    'return_images' => false,          // Include image results
    'search_recency_filter' => 'month' // Filter by time (hour, day, week, month, year)
]);

$data = json_decode($response->getBody()->getContents(), true);

// Access response with metadata
echo "Content: " . $data['choices'][0]['message']['content'] . "\n";

// Access citations if available
if (isset($data['citations'])) {
    echo "\nCitations:\n";
    foreach ($data['citations'] as $citation) {
        echo "- " . $citation . "\n";
    }
}

```

--------------------------------

### Create Chat Completion with PHP Perplexity AI SDK

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

This PHP code snippet illustrates how to create a synchronous chat completion request using the Perplexity AI SDK. It includes sending messages, setting parameters like model and temperature, and handling the response, including potential API errors.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAI;
use SoftCreatR\PerplexityAI\Exception\PerplexityAIException;

try {
    // Make a chat completion request
    $response = $pplx->createChatCompletion([], [
        'model' => 'llama-3.1-sonar-small-128k-online',
        'messages' => [
            [
                'role' => 'system',
                'content' => 'Be precise and concise.'
            ],
            [
                'role' => 'user',
                'content' => 'How many stars are there in our galaxy?'
            ]
        ],
        'max_tokens' => 150,
        'temperature' => 0.7
    ]);

    // Process successful response
    if ($response->getStatusCode() === 200) {
        $data = json_decode($response->getBody()->getContents(), true);

        // Extract the assistant's message
        $assistantMessage = $data['choices'][0]['message']['content'];
        echo "Response: " . $assistantMessage . "\n";

        // Access usage information
        echo "Tokens used: " . $data['usage']['total_tokens'] . "\n";
    }
} catch (PerplexityAIException $e) {
    echo "API Error: " . $e->getMessage() . "\n";
    echo "Status Code: " . $e->getCode() . "\n";
}

```

--------------------------------

### Implement Multi-turn Conversations with Perplexity AI SDK (PHP)

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Shows how to manage conversation history for multi-turn interactions with the Perplexity AI chat completion endpoint. This involves sending initial messages, receiving assistant responses, and including them in subsequent requests to maintain context.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAI;

// Maintain conversation history
$conversationHistory = [
    [
        'role' => 'system',
        'content' => 'You are a helpful coding assistant.'
    ],
    [
        'role' => 'user',
        'content' => 'Write a PHP function to calculate factorial.'
    ]
];

// First turn
$response1 = $pplx->createChatCompletion([], [
    'model' => 'llama-3.1-sonar-small-128k-online',
    'messages' => $conversationHistory
]);

$data1 = json_decode($response1->getBody()->getContents(), true);
$assistantReply1 = $data1['choices'][0]['message'];

// Add assistant's response to history
$conversationHistory[] = $assistantReply1;

// Add follow-up question
$conversationHistory[] = [
    'role' => 'user',
    'content' => 'Now add error handling for negative numbers.'
];

// Second turn
$response2 = $pplx->createChatCompletion([], [
    'model' => 'llama-3.1-sonar-small-128k-online',
    'messages' => $conversationHistory
]);

$data2 = json_decode($response2->getBody()->getContents(), true);
echo $data2['choices'][0]['message']['content'];

```

--------------------------------

### Create Chat Completion using PerplexityAI SDK (PHP)

Source: https://github.com/softcreatr/php-perplexity-ai-sdk/blob/main/README.md

Shows how to create a chat completion using the PerplexityAI PHP SDK. It sends a request with model and message details and processes the response, checking for a 200 status code before decoding the JSON body.

```php
$response = $pplx->createChatCompletion([], 
    [
        'model' => 'sonar',
        'messages' => [
            [
                'role' => 'system',
                'content' => 'Be precise and concise.'
            ],
            [
                'role' => 'user',
            '   content' => 'How many stars are there in our galaxy?'
        ]
    ],
]);

// Process the API response
if ($response->getStatusCode() === 200) {
    $responseObj = json_decode($response->getBody()->getContents(), true);
    
    print_r($responseObj);
} else {
    echo "Error: " . $response->getStatusCode();
}

```

--------------------------------

### Stream Chat Completion Responses with PerplexityAI SDK (PHP)

Source: https://github.com/softcreatr/php-perplexity-ai-sdk/blob/main/README.md

Illustrates how to enable and handle streaming responses for chat completions using the PerplexityAI PHP SDK. A callback function processes incoming data chunks, echoing the content from the 'choices' array.

```php
$streamCallback = static function ($data) {
    if (isset($data['choices'][0]['delta']['content'])) {
        echo $data['choices'][0]['delta']['content'];
    }
};

$pplx->createChatCompletion([],
    [
        'model' => 'sonar',
        'messages' => [
            [
                'role' => 'user',
                'content' => 'Tell me a story about a brave knight.',
            ],
        ],
        'stream' => true,
    ],
    $streamCallback
);

```

--------------------------------

### Handle Perplexity AI SDK Errors Gracefully (PHP)

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

Provides a comprehensive error handling structure for the PerplexityAI SDK, catching specific exceptions like PerplexityAIException, ClientExceptionInterface, InvalidArgumentException, and JsonException. This ensures robust handling of API, network, and data processing issues.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAI;
use SoftCreatR\PerplexityAI\Exception\PerplexityAIException;
use Psr\Http\Client\ClientExceptionInterface;

try {
    $response = $pplx->createChatCompletion([], [
        'model' => 'invalid-model-name',
        'messages' => [
            ['role' => 'user', 'content' => 'Test']
        ]
    ]);

    $data = json_decode($response->getBody()->getContents(), true);

} catch (PerplexityAIException $e) {
    // Handles API-level errors (4xx, 5xx responses)
    echo "API Error: " . $e->getMessage() . "\n";
    echo "HTTP Status: " . $e->getCode() . "\n";

    // Check for previous exception
    if ($previous = $e->getPrevious()) {
        echo "Underlying cause: " . $previous->getMessage() . "\n";
    }

} catch (ClientExceptionInterface $e) {
    // Handles HTTP client errors (network issues, timeouts)
    echo "HTTP Client Error: " . $e->getMessage() . "\n";

} catch (\InvalidArgumentException $e) {
    // Handles invalid parameters or configuration
    echo "Invalid Argument: " . $e->getMessage() . "\n";

} catch (\JsonException $e) {
    // Handles JSON encoding/decoding errors
    echo "JSON Error: " . $e->getMessage() . "\n";
}

```

--------------------------------

### Stream Chat Completion with PHP Perplexity AI SDK

Source: https://context7.com/softcreatr/php-perplexity-ai-sdk/llms.txt

This PHP code snippet demonstrates how to create a streaming chat completion request using the Perplexity AI SDK. It defines a callback function to process response chunks in real-time and handles potential streaming errors.

```php
<?php

use SoftCreatR\PerplexityAI\PerplexityAI;
use SoftCreatR\PerplexityAI\Exception\PerplexityAIException;

try {
    // Define streaming callback to process real-time chunks
    $streamCallback = static function (array $data): void {
        // Extract and display content as it arrives
        if (isset($data['choices'][0]['delta']['content'])) {
            echo $data['choices'][0]['delta']['content'];
            flush();
        }

        // Handle stream metadata
        if (isset($data['choices'][0]['finish_reason'])) {
            echo "\n\nStream finished: " . $data['choices'][0]['finish_reason'] . "\n";
        }
    };

    // Create streaming chat completion
    $pplx->createChatCompletion(
        [],
        [
            'model' => 'llama-3.1-sonar-small-128k-online',
            'messages' => [
                [
                    'role' => 'user',
                    'content' => 'Tell me a story about a brave knight in three paragraphs.'
                ]
            ],
            'stream' => true,
            'temperature' => 0.8
        ],
        $streamCallback
    );

    // Note: Returns null for streaming requests
} catch (PerplexityAIException $e) {
    echo "Streaming Error: " . $e->getMessage() . "\n";
}

```

=== COMPLETE CONTENT === This response contains all available snippets from this library. No additional content exists. Do not make further requests.