README

使用 laragpt 获取 OpenAI API 的力量。

laragpt 是开发者通过易于使用的 Laravel 接口访问 OpenAI 强大人工智能能力的理想包。使用 laragpt，您可以快速启动并运行 OpenAI 强大的深度学习算法、自然语言处理和机器学习能力，以便开发创新和复杂的应用程序。通过利用 OpenAI API 和 Laravel 简单的接口，laragpt 使将 AI 集成到项目中变得简单。使用 laragpt 给您的应用程序带来 OpenAI 的智能。

安装

您可以通过 composer 安装此包

composer require jumbaeric/laragpt

发布配置文件

php artisan vendor:publish

并选择发布 laragpt 包。

将 openai api 密钥添加到您的环境变量中

OPENAI_API_KEY=

用法

所有方法返回数组

use Jumbaeric\Laragpt\Laragpt;

// Complete
// Given a prompt, the model will return one or more predicted completions, and can also return the probabilities of alternative tokens at each position.
    $args = [
        'prompt' => 'Brainstorm some ideas combining VR and fitness', //required
        'model' => 'text-davinci-003',  //  required
    ];

    Laragpt::complete($args);

// Chat
// Given a chat conversation, the model will return a chat completion response.

Laragpt::chat([
        'model' => 'gpt-3.5-turbo', // required
        'messages' => [             // required
            [
                "role" => "system",
                "content" => "You are a helpful assistant."
            ],
            [
                "role" => "user",
                "content" => "Who won the world series in 2020?"
            ],
            [
                "role" => "assistant",
                "content" => "The Los Angeles Dodgers won the World Series in 2020."
            ],
            [
                "role" => "user",
                "content" => "Where was it played?"
            ],
        ],
        'temperature' => 1.0,   //  optional
        'max_tokens' => 4000,   //  optional
        'frequency_penalty' => 0,   //  optional
        'presence_penalty' => 0,    //  optional
    ]);

// Audio
// trascribe: Transcribes audio into the input language.
// translate: Translates audio into into English.

$type = "transcript"; // string: Required : value can be transcript or translation
Laragpt::audio([
            'model' => 'whisper-1',     //  required
            'file' => "audio.mp3",      //  required
            'prompt' => "",             //  optional
            'response_format' => '',    //  optional
            'temperature' => '',        //  optional
        ], $type);

// Edits
// Given a prompt and an instruction, the model will return an edited version of the prompt.
Laragpt::edits([
        'model' => 'text-davinci-edit-001', // required
        'input' => 'What day of the wek is it?',
        'instruction' => 'Fix the spelling mistakes',   //  required
        'n' => 1,   //  optional
        'temperature' => 1,   //  optional
        'top_p' => 1,    //  optional
    ]);

// Images
    $createArr = [ // $type = create
        'prompt' => 'A cute baby sea otter',
        'size' => '1024x1024',   //  required
        'n' => 1,   //  optional
    ] ;

    $ceateEditArr = [ // $type = createEdit
        'prompt' => 'A cute baby sea otter wearing a beret',    // required
        'image' => '@otter.png',                                //required
        'mask' => '@mask.png',
        'n' => 1,
        'size' => '1024x1024',
    ];

    $createVariationArr = [ // $type = variations
        'image' => '@otter.png',
        'n' => 2,
        'size' => '1024x1024',
    ];
    
    Laragpt::images($createArr, $type = 'create');

// Models
$model_id = null; // String : Optional : Lists all models when value is null, model set returns the model details
Laragpt::models($model_id);

测试

composer test

变更日志

请参阅变更日志以获取更多关于最近更改的信息。

贡献

请参阅贡献指南以获取详细信息。

安全

如果您发现任何安全相关的问题，请通过jumbaeric@gmail.com发送电子邮件，而不是使用问题跟踪器。

鸣谢

• Eric Jumba
• 所有贡献者

许可证

MIT 许可证 (MIT)。有关更多信息，请参阅许可证文件。