README

Official PHP SDK for Apiframe - The ultimate platform for AI image and video generation APIs.

Features

• 🎨 Midjourney API (Original) - Generate, upscale, vary, blend, inpaint, face swap
• 🚀 Midjourney Pro API - Fast & Turbo modes, better stability
• ⚡ Flux AI - Fast and high-quality image generation
• 🎭 Ideogram - Creative image generation with text rendering
• 🎬 Luma AI - Text and image to video generation
• 🎵 Suno AI - AI music generation
• 🎶 Udio AI - Advanced music creation
• 🎥 Runway ML - Gen-3 video generation
• 🎪 Kling AI - Video generation and manipulation
• 📸 AI Photos - Headshots, face swap, and photo enhancement
• 📤 Media Upload - Upload and manage media files

Requirements

• PHP 7.4 or higher
• Composer
• An Apiframe API key (get one at apiframe.ai)

Installation

Install via Composer:

composer require apiframe/php-sdk

Quick Start

<?php

require 'vendor/autoload.php';

use Apiframe\Apiframe;

$client = new Apiframe([
    'apiKey' => 'your_api_key_here'
]);

// Create an image generation task
$task = $client->midjourney->imagine([
    'prompt' => 'a serene mountain landscape at sunset, photorealistic',
    'aspect_ratio' => '16:9'
]);

echo "Task created: {$task['id']}\n";

// Wait for completion with progress updates
$result = $client->tasks->waitFor($task['id'], [
    'onProgress' => function($progress) {
        echo "Progress: {$progress}%\n";
    }
]);

echo "Images ready: " . json_encode($result['image_urls']) . "\n";

API Reference

Configuration

$client = new Apiframe([
    'apiKey' => 'your_api_key',                    // Required: Your Apiframe API key
    'baseURL' => 'https://api.apiframe.ai',        // Optional: Custom API endpoint
    'timeout' => 300                                // Optional: Request timeout in seconds (default: 300)
]);

Midjourney (Original API)

The original Midjourney API with comprehensive features.

Endpoint: /imagine, /imagine-video, /reroll, /variations, /faceswap, etc.
Docs: https://docs.apiframe.ai/api-endpoints

imagine($params)

Create a new image generation task.

$task = $client->midjourney->imagine([
    'prompt' => 'a serene mountain landscape',
    'aspect_ratio' => '16:9',                      // Optional: '1:1', '16:9', '9:16', etc.
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

imagineVideo($params)

Generate videos using a text prompt and an image URL.

$task = $client->midjourney->imagineVideo([
    'prompt' => 'cinematic mountain landscape',
    'image_url' => 'https://example.com/start-frame.jpg',
    'motion' => 'high',                            // Optional: 'low' or 'high'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

reroll($params)

Reroll to create new images from a previous Imagine task.

$task = $client->midjourney->reroll([
    'parent_task_id' => 'original_task_id',
    'prompt' => 'optional new prompt',             // Optional: uses original prompt if not provided
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

variations($params)

Create 4 new variations of one of the 4 generated images.

$task = $client->midjourney->variations([
    'parent_task_id' => 'original_task_id',
    'index' => '1',                                // '1', '2', '3', '4', or 'strong', 'subtle'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

faceSwap($params)

Swap the face on a target image with the face on a provided image.

$task = $client->midjourney->faceSwap([
    'target_image_url' => 'https://example.com/target.jpg',
    'swap_image_url' => 'https://example.com/face.jpg',
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

upscale1x($params)

Upscale one of the 4 generated images to get a single image.

$task = $client->midjourney->upscale1x([
    'parent_task_id' => 'original_task_id',
    'index' => '1',                                // '1', '2', '3', or '4'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

upscaleAlt($params)

Upscale with Subtle or Creative mode.

$task = $client->midjourney->upscaleAlt([
    'parent_task_id' => 'upscale1x_task_id',
    'type' => 'subtle',                            // 'subtle' or 'creative'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

upscaleHighres($params)

Upscale any image to higher resolution (2x or 4x).

$task = $client->midjourney->upscaleHighres([
    'parent_task_id' => 'task_id',                 // Or use image_url instead
    'image_url' => 'https://example.com/image.jpg', // Or use parent_task_id instead
    'type' => '2x',                                // '2x' or '4x'
    'index' => '1',                                // Optional: '1', '2', '3', or '4'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

inpaint($params)

Redraw a selected area of an image (Vary Region).

$task = $client->midjourney->inpaint([
    'parent_task_id' => 'upscale1x_task_id',
    'mask' => 'base64_encoded_mask_image',
    'prompt' => 'a red sports car',
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

outpaint($params)

Enlarges an image's canvas beyond its original size.

$task = $client->midjourney->outpaint([
    'parent_task_id' => 'upscale1x_task_id',
    'zoom_ratio' => 1.5,                           // Can be 1.5, 2, or 1 for custom zoom
    'aspect_ratio' => '1:1',                       // Optional
    'prompt' => 'mountain landscape',              // Optional
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

pan($params)

Broadens the image canvas in a specific direction.

$task = $client->midjourney->pan([
    'parent_task_id' => 'upscale1x_task_id',
    'direction' => 'up',                           // 'up', 'down', 'left', or 'right'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

describe($params)

Writes four example prompts based on an image.

$task = $client->midjourney->describe([
    'image_url' => 'https://example.com/image.jpg',
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

blend($params)

Blend multiple images into one image (2-5 images).

$task = $client->midjourney->blend([
    'image_urls' => [
        'https://example.com/image1.jpg',
        'https://example.com/image2.jpg'
    ],
    'dimension' => 'square',                       // Optional: 'square', 'portrait', or 'landscape'
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

shorten($params)

Analyzes your prompt and suggests optimizations.

$task = $client->midjourney->shorten([
    'prompt' => 'a very beautiful and amazing sunset over the mountains with clouds',
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

seed($params)

Get the seed of a generated image.

$task = $client->midjourney->seed([
    'task_id' => 'original_task_id',
    'webhook_url' => 'https://your-domain.com/webhook',    // Optional
    'webhook_secret' => 'your-secret'                       // Optional
]);

Midjourney Pro API (MidjourneyAlt)

The Pro Midjourney API with Fast & Turbo modes.

Endpoint: /pro/midjourney/*
Docs: https://docs.apiframe.ai/pro-midjourney-api/api-endpoints

imagine($params)

Create a new image with Pro API.

$task = $client->midjourneyAlt->imagine([
    'prompt' => 'a serene mountain landscape',
    'mode' => 'turbo'                              // 'fast' or 'turbo'
]);

upscale($params)

Upscale a specific image from a Pro API generation.

$task = $client->midjourneyAlt->upscale([
    'parent_task_id' => 'parent_task_id',
    'index' => '1',                                // '1', '2', '3', or '4'
    'type' => 'subtle'                             // 'subtle' or 'creative'
]);

vary($params)

Create variations with strong/subtle control.

$task = $client->midjourneyAlt->vary([
    'parent_task_id' => 'parent_task_id',
    'index' => '1',                                // '1', '2', '3', or '4'
    'type' => 'subtle'                             // 'subtle' or 'strong'
]);

Flux

// Generate with Flux (specify model)
$task = $client->flux->generate([
    'model' => 'flux-pro',                         // 'flux-schnell', 'flux-pro', 'flux-dev', etc.
    'prompt' => 'a futuristic cityscape',
    'width' => 1024,
    'height' => 1024,
    'steps' => 50,                                 // only for flux-pro and flux-dev
    'guidance' => 7.5,                             // only for flux-pro and flux-dev
    'seed' => 42,
    'safety_tolerance' => 2
]);

// Convenience methods (automatically set model)
$task = $client->flux->generatePro([
    'prompt' => 'a futuristic cityscape',
    'width' => 1024,
    'height' => 1024
]);

$task = $client->flux->generateDev([
    'prompt' => 'a landscape',
    'aspect_ratio' => '16:9'
]);

$task = $client->flux->generateSchnell([
    'prompt' => 'quick sketch',
    'width' => 512,
    'height' => 512
]);

Ideogram

// Generate image
$task = $client->ideogram->generate([
    'prompt' => 'a logo design',
    'aspect_ratio' => 'ASPECT_1_1',
    'style_type' => 'DESIGN',
    'magic_prompt_option' => 'AUTO',
    'seed' => 12345,
    'resolution' => 'RESOLUTION_1024_1024'
]);

// Upscale image
$task = $client->ideogram->upscale([
    'image_url' => 'https://...',
    'prompt' => 'enhance this image',
    'resemblance' => 80,                           // 1-100
    'detail' => 50,                                // 1-100
    'seed' => 12345
]);

// Describe image
$task = $client->ideogram->describe([
    'image_url' => 'https://...'
]);

// Remix (image-to-image)
$task = $client->ideogram->remix([
    'image_url' => 'https://...',
    'prompt' => 'transform this image...',
    'image_weight' => 70,                          // 1-100
    'style_type' => 'REALISTIC'
]);

Luma AI (Video)

// Generate video from text prompt
$task = $client->luma->generate([
    'prompt' => 'a serene beach with waves',
    'aspect_ratio' => '16:9',
    'loop' => false,
    'enhance_prompt' => true
]);

// Generate video with start and end images
$task = $client->luma->generate([
    'prompt' => 'a smooth transition',
    'image_url' => 'https://start-image.jpg',
    'end_image_url' => 'https://end-image.jpg',
    'aspect_ratio' => '1:1'
]);

// Extend a previously generated video
$task = $client->luma->extend([
    'parent_task_id' => 'previous_task_id',
    'prompt' => 'continue the scene with more action'
]);

Suno AI (Music)

// Generate song with lyrics (creates TWO songs)
$task = $client->suno->generate([
    'prompt' => 'an upbeat electronic track',
    'lyrics' => 'Verse 1: Dancing through the night...',
    'model' => 'chirp-v3-5',
    'tags' => 'electronic, dance, upbeat',
    'title' => 'Digital Dreams',
    'make_instrumental' => false
]);

// Result will contain TWO songs with same lyrics
$result = $client->tasks->waitFor($task['id']);
// $result['songs'] will contain array of 2 songs

// Upload audio and turn it into extendable song
$uploadTask = $client->suno->upload([
    'audio_url' => 'https://your-audio-url.mp3'
]);
$uploadResult = $client->tasks->waitFor($uploadTask['id']);

// Extend a song
$extendTask = $client->suno->extend([
    'song_id' => $uploadResult['song_id'],
    'continue_at' => 30,
    'from_upload' => true,
    'prompt' => 'continue with more energy'
]);

// Generate lyrics only
$lyricsTask = $client->suno->generateLyrics([
    'prompt' => 'a song about summer adventures'
]);
$lyricsResult = $client->tasks->waitFor($lyricsTask['id']);

Udio AI (Music)

// Generate music (creates TWO songs with lyrics)
$task = $client->udio->generate([
    'prompt' => 'a calm ambient soundtrack',
    'lyrics' => 'Verse 1: Under the stars...',
    'model' => 'udio32-v1.5',
    'tags' => 'ambient, calm, instrumental',
    'prompt_strength' => 0.8,
    'clarity_strength' => 0.7,
    'lyrics_strength' => 0.6,
    'generation_quality' => 0.9,
    'lyrics_placement_start' => 4,
    'lyrics_placement_end' => 20,
    'bypass_prompt_optimization' => false
]);

$result = $client->tasks->waitFor($task['id']);
// $result['songs'] will contain array of 2 songs

Runway ML (Video)

// Generate video
$task = $client->runway->generate([
    'prompt' => 'a drone shot flying over mountains',
    'generation_type' => 'text2video',
    'model' => 'gen3',
    'aspect_ratio' => '16:9',
    'duration' => 10,
    'flip' => false
]);

// Convenience method: Text to video
$task = $client->runway->textToVideo(
    'a drone shot flying over mountains',
    ['model' => 'gen3a_turbo', 'duration' => 5]
);

// Convenience method: Image to video
$task = $client->runway->imageToVideo(
    'https://image-url.jpg',
    'add cinematic motion to this scene',
    ['duration' => 10]
);

// Convenience method: Video to video
$task = $client->runway->videoToVideo(
    'https://video-url.mp4',
    'transform this video with a sunset atmosphere',
    ['model' => 'gen3', 'duration' => 5]
);

Kling AI (Video)

// Generate video
$task = $client->kling->generate([
    'prompt' => 'a time-lapse of a flower blooming',
    'generation_type' => 'text2video',
    'model' => 'kling-v1-5',
    'mode' => 'pro',
    'aspect_ratio' => '16:9',
    'duration' => 10,
    'cfg_scale' => 0.5
]);

// Convenience method: Text to video
$task = $client->kling->textToVideo(
    'a time-lapse of a flower blooming',
    ['duration' => 10, 'aspect_ratio' => '16:9']
);

// Convenience method: Image to video
$task = $client->kling->imageToVideo(
    'https://image-url.jpg',
    'animate this image with smooth motion',
    ['mode' => 'pro', 'duration' => 5]
);

// Virtual Try On
$task = $client->kling->tryon([
    'human_image_url' => 'https://person-image.jpg',
    'cloth_image_url' => 'https://clothing-image.jpg',
    'model' => 'kolors-virtual-try-on-v1-5'
]);

AI Photos

// Step 1: Upload and prepare 10-30 images for training
$uploadTask = $client->aiPhotos->upload([
    'images' => ['base64_image_1', 'base64_image_2', '...'], // 10-30 images
    'ethnicity' => 'white',
    'gender' => 'male',
    'age' => 30
]);

$uploadResult = $client->tasks->waitFor($uploadTask['id']);
echo "Images ready for training\n";

// Step 2: Train AI on the subject
$trainTask = $client->aiPhotos->train([
    'training_images_id' => $uploadTask['id'],
    'trigger_word' => 'TOKMSN'                     // Default trigger word
]);

$trainResult = $client->tasks->waitFor($trainTask['id']);
echo "Training finished, trigger_word: {$trainResult['trigger_word']}\n";

// Step 3: Generate photos using the trained model
$generateTask = $client->aiPhotos->generate([
    'training_id' => $trainTask['id'],
    'prompt' => 'a realistic portrait of TOKMSN black man wearing a suit',
    'aspect_ratio' => '1:1',
    'number_of_images' => '4',
    'seed' => 12345
]);

$result = $client->tasks->waitFor($generateTask['id']);
echo "Generated photos: " . json_encode($result['image_urls']) . "\n";

Media Upload

// Upload image from file (max 2MB)
$upload = $client->media->upload([
    'filename' => './path/to/image.jpg'
]);

echo "Uploaded: {$upload['imageURL']}\n";

// Upload audio from file (max 2MB, 60 seconds)
$audioUpload = $client->media->uploadAudio([
    'filename' => './path/to/audio.mp3'
]);

echo "Uploaded audio: {$audioUpload['audioURL']}\n";

// Use uploaded media
$task = $client->midjourney->blend([
    'image_urls' => [$upload['imageURL'], $another_url]
]);

Tasks

General task management endpoints.

get($taskId)

Get the result/status of a submitted task.

$task = $client->tasks->get($taskId);
echo "Status: {$task['status']}\n"; // 'pending', 'processing', 'completed', 'failed'

getMany($taskIds)

Get the results/statuses of multiple tasks (min 2, max 20).

$result = $client->tasks->getMany(['task_id_1', 'task_id_2', 'task_id_3']);
echo "Tasks: " . count($result['tasks']) . "\n";

waitFor($taskId, $options)

Wait for a task to complete with progress tracking.

$result = $client->tasks->waitFor($taskId, [
    'onProgress' => function($progress) {
        echo "Progress: {$progress}%\n";
    },
    'interval' => 3000,                            // Polling interval in ms (default: 3000)
    'timeout' => 300000                            // Max wait time in ms (default: 300000 / 5 min)
]);

getAccountInfo()

Get account details including credits, usage, plan, etc.

$account = $client->tasks->getAccountInfo();
echo "Email: {$account['email']}\n";
echo "Credits: {$account['credits']}\n";
echo "Total Images: {$account['total_images']}\n";
echo "Plan: {$account['plan']}\n";

Error Handling

use Apiframe\Exceptions\ApiframeException;
use Apiframe\Exceptions\AuthenticationException;
use Apiframe\Exceptions\RateLimitException;
use Apiframe\Exceptions\TimeoutException;

try {
    $task = $client->midjourney->imagine(['prompt' => '...']);
    $result = $client->tasks->waitFor($task['id']);
} catch (AuthenticationException $e) {
    echo "Invalid API key\n";
} catch (RateLimitException $e) {
    echo "Rate limit exceeded\n";
} catch (TimeoutException $e) {
    echo "Request timed out\n";
} catch (ApiframeException $e) {
    echo "API error: {$e->getMessage()}\n";
    echo "Status: {$e->getStatus()}\n";
}

Environment Variables

You can set your API key using an environment variable:

export APIFRAME_API_KEY=your_api_key_here

Then in your code:

$client = new Apiframe([
    'apiKey' => getenv('APIFRAME_API_KEY')
]);

Examples

Check the /examples directory for complete examples:

• midjourney-example.php - Midjourney Original API usage
• midjourney-alt-example.php - Midjourney Pro API usage
• flux-example.php - Flux AI image generation
• luma-video-example.php - Luma video generation
• suno-music-example.php - Suno music generation
• media-upload-example.php - Media upload and usage
• comprehensive-example.php - All features demo

Documentation

For detailed API documentation, visit docs.apiframe.ai

Support

• Documentation: docs.apiframe.ai
• GitHub Issues: github.com/apiframe/apiframe-php-sdk/issues

License

MIT License - see LICENSE file for details.

Contributing

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