OpenAI Models

The Luminova OpenAI class provides an easy to use methods for integrating OpenAI's powerful language models into your applications. This class abstracts the complexities of interacting with OpenAI's API, enabling developers to effortlessly harness the capabilities of models like ChatGPT, DALL-E, Whisper, and others. Whether you're building chatbots, generating images, performing text analysis, or converting speech, the OpenAI class offers a useful methods to embedding advanced AI functionalities into your projects. To learn more about OpenAI Integration.

Example Usage

Here's a basic example demonstrating how to use the Luminova\OpenAI class to generate text completions:

use \Luminova\Ai\Models\OpenAI;
use \Luminova\Exceptions\AppException;

$openai = new OpenAI('your-api-key');

try {
    $options = ['model' => 'gpt-3.5-turbo-instruct'];
    $response = $openai->completion('Tell me a joke.', $options);
    print_r($response);
} catch (AppException $e) {
    echo 'Error: ' . $e->getMessage();
}

Using the AI Model class for easy implementation, this is useful when you want to switch AIs.

use \Luminova\Ai\Model;
use \Luminova\Ai\Models\OpenAI;
use \Luminova\Exceptions\AppException;

$ai = new Model(); // Optional pass AI class instance (default: OpenAI)

try {
    $options = ['model' => 'gpt-3.5-turbo-instruct'];
    $response = $ai->completion('Tell me a joke.', $options);
    print_r($response);
} catch (AppException $e) {
    echo 'Error: ' . $e->getMessage();
}

• Class namespace: \Luminova\Ai\Models\OpenAI
• This class implements:\Luminova\Interface\AiInterface

Methods

To get started, Set up your API key, and optional organization and project IDs.

constructor

Constructs a new instance.

public __construct(string $apiKey, string $version = 'v1', string|null $organization = null, string|null $project = null)

Parameters:

completion

Get completions for the given prompt.

public completion(string $prompt, array $options = []): array

Parameters:

Return Value:array - Returns an array of completion choices.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

suggestions

Get suggestions for the given prompt.

public suggestions(string $prompt, array $options = []): array

Parameters:

Return Value:array - Returns an array of suggestion choices.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

message

Send a message and receive replies.

public message(string $prompt, array $options = []): array

Parameters:

Return Value:array - Returns an array of replies.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

image

Generate random images.

public image(string $prompt, array $options = []): array|false

Parameters:

Return Value:array|false - An array of image URLs or base64 encoded images, false on failure.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

imageEdit

Edit or swap areas of an image with another image.

public imageEdit(string $prompt, array $options): array|false

Parameters:

Return Value:array|false - An array of edited image URLs or base64 encoded images, false on failure.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

fineTune

Create additional training on how ChatGPT should behave based on your training examples.

public fineTune(string $trainingFile, array $options = []): array

Parameters:

Return Value:array - Returns an array of training responses.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

embed

Get text embeddings to measure the relatedness of text strings.

public embed(string|array $input, array $options = []): array

Parameters:

Return Value:array - Returns an array of embedding vectors.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

speech

Generate audio from text.

public speech(string $text, array $options = []): string|false

Parameters:

Return Value:string|false - Return file url on success, false on failure.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

audio

Translates audio to text.

public audio(string $filename, array $options = []): string|false

Parameters:

Return Value:string|false - Returns the converted audio on success, false on failure.

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

models

Retrieve a list of available models.

public models(string|null $name = null): array

Parameters:

Return Value:array - An array containing information about the model(s).

Throws:

• \Luminova\Exceptions\RuntimeException - If an error is encountered during the network request.
• \Luminova\Exceptions\JsonException - If an error is encountered during JSON decoding.

API Options

Shared Options

• model (string) (required) - The model to use for the API call.
• n (int) (optional) - The number of items to generate.
• user (string) (optional) - The identifier for the end-user.

Conversions

• max_tokens (int) (optional) - The maximum number of tokens to generate.
• temperature (float) (optional) - The sampling temperature to use, ranging from 0 to 1.

Image Generation

• size (string) (optional) - The dimensions of the generated image, e.g., 256x256.
• response_format (string) (optional) - The format of the API response, e.g., url or base64_json.

Image Editing

• image (string) (required) - The path to the original image to edit (PNG only, max 4MB).
• mask (string) (required) - The path to the image to use as a mask (PNG only, max 4MB).
• size (string) (optional) - The dimensions of the edited image, e.g., 256x256.
• response_format (string) (optional) - The format of the API response, e.g., url or base64_json.

Speech Generation

• voice (string) (optional) - The voice to use for audio conversion. Available voices: alloy, echo, fable, onyx, nova, or shimmer. Default is alloy.
• path (string) (optional) - The destination path to save the converted file. Default is writeable/ai/speech.
• symlink (string) (optional) - The optional symbolic link destination for audio files.
• response_format (string) (optional) - The format of the API response, e.g., mp3, wav.

Audio Transcription

• prompt (string) (optional) - The optional text prompt to guide the transcription.
• language (string) (optional) - The language of the audio file.
• response_format (string) (optional) - The format of the API response, e.g., json, text.
• temperature (float) (optional) - The sampling temperature to use for transcription, ranging from 0 to 1.

Fine-Tuning

• validation_file (string) (required) - The path to the validation file for fine-tuning.
• training_file (string) (required) - The path to the training file for fine-tuning.

Embedding

• dimensions (int) (optional) - The number of dimensions for the embedding vector.
• encoding_format (string) (optional) - The format of the embedding, e.g., json, binary.