ChatGPT API Wrapper

A simple and lightweight npm package for interacting with OpenAI's ChatGPT API, supporting both text completion and image generation.

Features

• 🤖 Text Completion: Support for all ChatGPT models (GPT-3.5, GPT-4, etc.)
• 🎨 Image Generation: DALL-E 2 and DALL-E 3 support
• 🔧 TypeScript Support: Full type definitions included
• ⚡ Simple API: Easy-to-use methods for common tasks
• 🛡️ Error Handling: Comprehensive error handling and validation
• 🔄 Flexible Configuration: Customizable timeout, base URL, and model selection

Installation

npm install chatgpt-api-wrapper

Quick Start

const ChatGPT = require('chatgpt-api-wrapper');

const client = new ChatGPT({
  apiKey: 'your-openai-api-key'
});

// Simple text completion
const response = await client.complete('What is artificial intelligence?');
console.log(response);

// Generate an image
const image = await client.generateImage({
  prompt: 'A beautiful sunset over mountains',
  model: 'dall-e-3'
});
console.log(image.data[0].url);

API Reference

Constructor

const client = new ChatGPT({
  apiKey: 'your-api-key',        // Required
  baseURL: 'custom-base-url',    // Optional
  timeout: 30000                 // Optional (default: 30000ms)
});

Text Completion

Simple Completion

const response = await client.complete(prompt, model);

Advanced Chat

const response = await client.chat({
  model: ChatGPT.MODELS.GPT_4,
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Hello!' }
  ],
  maxTokens: 150,
  temperature: 0.7,
  topP: 1,
  frequencyPenalty: 0,
  presencePenalty: 0
});

Image Generation

const image = await client.generateImage({
  prompt: 'A cute robot',
  model: 'dall-e-3',          // 'dall-e-2' or 'dall-e-3'
  size: '1024x1024',          // Various sizes available
  quality: 'standard',        // 'standard' or 'hd' (DALL-E 3 only)
  style: 'vivid',             // 'vivid' or 'natural' (DALL-E 3 only)
  n: 1                        // Number of images
});

Available Models

console.log(ChatGPT.MODELS);
// Output:
// {
//   GPT_4: 'gpt-4',
//   GPT_4_TURBO: 'gpt-4-turbo',
//   GPT_4_TURBO_PREVIEW: 'gpt-4-turbo-preview',
//   GPT_3_5_TURBO: 'gpt-3.5-turbo',
//   GPT_3_5_TURBO_16K: 'gpt-3.5-turbo-16k'
// }

Utility Methods

// Test API connection
const isConnected = await client.testConnection();

// Get available models
const models = client.getAvailableModels();

Configuration Options

| Option | Type | Default | Description | |--------|------|---------|-------------| | apiKey | string | Required | Your OpenAI API key | | baseURL | string | https://api.openai.com/v1 | API base URL | | timeout | number | 30000 | Request timeout in milliseconds |

Image Generation Options

DALL-E 2

• Sizes: 256x256, 512x512, 1024x1024
• Up to 10 images per request

DALL-E 3

• Sizes: 1024x1024, 1792x1024, 1024x1792
• Quality: standard, hd
• Style: vivid, natural
• 1 image per request

Error Handling

The package provides detailed error messages for common issues:

try {
  const response = await client.complete('Hello');
} catch (error) {
  console.error('Error:', error.message);
  // Detailed error information from OpenAI API
}

Local Development & Testing

1. Clone this repository
2. Install dependencies: npm install
3. Build the package: npm run build
4. Set your API key: export OPENAI_API_KEY="your-key-here"
5. Run tests: npm test

TypeScript Support

This package includes full TypeScript definitions:

import ChatGPT, { ChatMessage, ChatCompletionOptions } from 'chatgpt-api-wrapper';

const client = new ChatGPT({ apiKey: 'your-key' });

const messages: ChatMessage[] = [
  { role: 'user', content: 'Hello!' }
];

const options: ChatCompletionOptions = {
  model: ChatGPT.MODELS.GPT_4,
  messages,
  temperature: 0.7
};

Examples

Chatbot with Context

const conversation = [
  { role: 'system', content: 'You are a helpful coding assistant.' },
  { role: 'user', content: 'How do I create a REST API in Node.js?' }
];

const response = await client.chat({
  model: ChatGPT.MODELS.GPT_4,
  messages: conversation,
  maxTokens: 500
});

// Add assistant's response to conversation
conversation.push({
  role: 'assistant',
  content: response.choices[0].message.content
});

Batch Image Generation

const prompts = [
  'A cat wearing a space suit',
  'A robot playing piano',
  'A mountain landscape at dawn'
];

const images = await Promise.all(
  prompts.map(prompt => 
    client.generateImage({ prompt, model: 'dall-e-2' })
  )
);

images.forEach((img, i) => {
  console.log(`Image ${i + 1}: ${img.data[0].url}`);
});

License

MIT

Contributing

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