README

A comprehensive PHP SDK for the Straico API - a unified platform for generative AI capabilities including text generation, image creation, chat, RAG (Retrieval Augmented Generation), and autonomous agents.

Features

Simple and intuitive API interface
Support for all Straico API endpoints (v0, v1, v2)
Environment variable configuration
PSR-4 autoloading
Comprehensive error handling
Full type hinting
File upload support
Multi-version API support

Requirements

PHP 7.4 or higher
Composer

Installation

Install via Composer:

composer require r5dy1n/straico-php

Configuration

API Key

Get your API key from Straico. You can provide it in two ways:

Environment Variable (recommended):

export STRAICO_API_KEY="your-api-key-here"

Constructor Parameter:

$straico = new \Straico\Straico('your-api-key-here');

Usage

Initialize Client

<?php

require_once 'vendor/autoload.php';

use Straico\Straico;
use Straico\Exceptions\StraicoException;

// Using environment variable
$straico = new Straico();

// Or with API key parameter
$straico = new Straico('your-api-key-here');

User Information

Get current user account details including coins, plan, and name:

try {
    $user = $straico->user();
    print_r($user);
} catch (StraicoException $e) {
    echo "Error: " . $e->getMessage();
}

Available Models

Retrieve list of available AI models with pricing and specifications:

// Get models (v0)
$models = $straico->models();

// Get models (v1)
$models = $straico->modelsV1();

// Get models (v2) - OpenAI-Compatible format
$models = $straico->modelsV2();

foreach ($models['data'] as $model) {
    echo $model['name'] . " - " . $model['pricing']['coins'] . " coins\n";
}

Get Specific Model

Get detailed information about a specific model:

$model = $straico->model('openai/gpt-4');
print_r($model);

Prompt Completion (Text Generation)

Generate text using AI models:

// Using v0 (default)
$response = $straico->promptCompletion(
    model: 'openai/gpt-4',
    message: 'Explain quantum computing in simple terms',
    options: []
);

// Using v1
$response = $straico->promptCompletionV1(
    model: 'openai/gpt-4',
    message: 'Explain quantum computing in simple terms',
    options: []
);

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

With Additional Options

$response = $straico->promptCompletion(
    model: 'openai/gpt-4',
    message: 'Analyze this image and describe what you see',
    options: [
        'images' => ['path/to/image.jpg'],
        'display_transcripts' => true
    ]
);

With YouTube URL Analysis

$response = $straico->promptCompletion(
    model: 'anthropic/claude-3-opus',
    message: 'Summarize this video',
    options: [
        'youtube_urls' => ['https://youtube.com/watch?v=...'],
        'display_transcripts' => true
    ]
);

Chat Completion

Use chat-based models with conversation history:

// Using v0 (default)
$response = $straico->chatCompletion(
    model: 'openai/gpt-4',
    messages: [
        ['role' => 'system', 'content' => 'You are a helpful assistant.'],
        ['role' => 'user', 'content' => 'What is the capital of France?']
    ]
);

// Using v1
$response = $straico->chatCompletionV1(
    model: 'openai/gpt-4',
    messages: [
        ['role' => 'user', 'content' => 'Hello!']
    ]
);

// Using v2
$response = $straico->chatCompletionV2(
    model: 'openai/gpt-4',
    messages: [
        ['role' => 'user', 'content' => 'Hello!']
    ],
    options: [
        'temperature' => 0.7,
        'max_tokens' => 150
    ]
);

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

File Upload

Upload files to use in other API calls:

$uploadResult = $straico->uploadFile('/path/to/document.pdf');
$fileId = $uploadResult['file_id'];
echo "File uploaded: $fileId\n";

Image Generation

Generate images from text descriptions:

$result = $straico->imageGeneration(
    model: 'openai/dall-e-3',
    description: 'A serene landscape with mountains and a lake at sunset',
    size: '1024x1024',
    variations: 1
);

$imageUrl = $result['data'][0]['url'];
echo "Generated image: $imageUrl\n";

RAG (Retrieval Augmented Generation)

Create RAG Instance

$rag = $straico->createRag(
    name: 'My Knowledge Base',
    description: 'Documentation and guides',
    files: [
        [
            'name' => 'document.pdf',
            'content' => base64_encode(file_get_contents('document.pdf'))
        ]
    ]
);

$ragId = $rag['id'];

List All RAGs

$rags = $straico->rags();

foreach ($rags as $rag) {
    echo $rag['name'] . " (ID: " . $rag['id'] . ")\n";
}

Get Specific RAG

$rag = $straico->rag('rag-id-here');
print_r($rag);

Update RAG with New Files

Add more files to an existing RAG:

$result = $straico->updateRag(
    ragId: 'rag-id-here',
    filePaths: [
        '/path/to/new-document.pdf',
        '/path/to/another-file.txt'
    ]
);

echo "RAG updated with " . $result['total_words'] . " words\n";

Query RAG with Prompt

// Basic query
$response = $straico->ragPromptCompletion(
    ragId: 'rag-id-here',
    model: 'openai/gpt-4',
    prompt: 'What does the documentation say about authentication?'
);

// With optional retriever parameters
$response = $straico->ragPromptCompletion(
    ragId: 'rag-id-here',
    model: 'openai/gpt-4',
    prompt: 'What does the documentation say about authentication?',
    options: [
        'search_type' => 'similarity',  // similarity, mmr, or similarity_score_threshold
        'k' => 4,                        // number of documents to return
        'fetch_k' => 20,                 // documents to pass to MMR algorithm
        'lambda_mult' => 0.5,            // diversity (0=max diversity, 1=min diversity)
        'score_threshold' => 0.7         // minimum relevance threshold
    ]
);

echo $response['data']['answer'];

// View references
foreach ($response['data']['references'] as $ref) {
    echo "Page {$ref['page']}: {$ref['page_content']}\n";
}

Delete RAG

$result = $straico->deleteRag('rag-id-here');

Agents

Create Agent

Create autonomous agents for task execution:

$agent = $straico->createAgent([
    'name' => 'Research Assistant',
    'description' => 'Helps with research tasks',
    'model' => 'openai/gpt-4',
    // Additional configuration...
]);

$agentId = $agent['_id'];

Query Agent

// Basic query
$response = $straico->agentPromptCompletion(
    agentId: 'agent-id-here',
    prompt: 'Research the latest developments in AI'
);

// With optional parameters
$response = $straico->agentPromptCompletion(
    agentId: 'agent-id-here',
    prompt: 'Research the latest developments in AI',
    options: [
        'search_type' => 'mmr',
        'k' => 5
    ]
);

echo $response['data']['answer'];

Error Handling

All methods throw StraicoException on errors:

use Straico\Exceptions\StraicoException;

try {
    $response = $straico->promptCompletion('openai/gpt-4', 'Hello');
} catch (StraicoException $e) {
    echo "Error: " . $e->getMessage() . "\n";
    echo "Code: " . $e->getCode() . "\n";
}

API Methods Reference

User

Method	Description	Version
`user()`	Get user account information	v0

Models

Method	Description	Version
`models()`	List available AI models	v0
`modelsV1()`	List available AI models	v1
`modelsV2()`	List available AI models (OpenAI-compatible)	v2
`model($modelId)`	Get specific model information	v2

Prompt Completion

Method	Description	Version
`promptCompletion($model, $message, $options)`	Generate text completion	v0
`promptCompletionV1($model, $message, $options)`	Generate text completion	v1

Chat Completion

Method	Description	Version
`chatCompletion($model, $messages, $options)`	Chat-based completion	v0
`chatCompletionV1($model, $messages, $options)`	Chat-based completion	v1
`chatCompletionV2($model, $messages, $options)`	Chat-based completion	v2

Files

Method	Description	Version
`uploadFile($filePath)`	Upload a file	v0

Images

Method	Description	Version
`imageGeneration($model, $description, $size, $variations)`	Generate images	v0

RAG

Method	Description	Version
`createRag($name, $description, $files)`	Create RAG instance	v0
`rags()`	List all RAGs	v0
`rag($ragId)`	Get specific RAG	v0
`updateRag($ragId, $filePaths)`	Update RAG with new files	v0
`deleteRag($ragId)`	Delete RAG	v0
`ragPromptCompletion($ragId, $model, $prompt, $options)`	Query RAG	v0

Agents

Method	Description	Version
`createAgent($config)`	Create agent	v0
`agentPromptCompletion($agentId, $prompt, $options)`	Query agent	v0

Official Documentation

For complete API documentation, visit:

License

MIT License

Support

For issues and questions:

GitHub Issues: Create an issue

Contributing

Contributions are welcome! Please feel free to submit a Pull Request.

r5dy1n / straico-php

Maintainers

Details

README

Features

Requirements

Installation

Configuration

API Key

Usage

Initialize Client

User Information

Available Models

Get Specific Model

Prompt Completion (Text Generation)

With Additional Options

With YouTube URL Analysis

Chat Completion

File Upload

Image Generation

RAG (Retrieval Augmented Generation)

Create RAG Instance

List All RAGs

Get Specific RAG

Update RAG with New Files

Query RAG with Prompt

Delete RAG

Agents

Create Agent

Query Agent

Error Handling

API Methods Reference

User

Models

Prompt Completion

Chat Completion

Files

Images

RAG

Agents

Official Documentation

License

Support

Contributing