aryai

A lightweight Node.js wrapper for Aryan Chauhan's AI APIs, providing easy access to multiple AI services including GPT-5, DeepSeek, Gemini, Llama, and more.

🚀 Features

• Multiple AI Models: Access to GPT-5, DeepSeek, Gemini, Llama-4, and more
• Simple API: Clean, promise-based interface
• Lightweight: Zero dependencies (uses Node.js built-in fetch)
• Image Analysis: Support for image analysis through Gemini Proxy
• Web Search: Brave search integration
• TypeScript Ready: Works great with TypeScript projects

📦 Installation

Using npm:

npm install aryai

Using yarn:

yarn add aryai

Using pnpm:

pnpm add aryai

🔧 Requirements

• Node.js 18+ (required for global fetch API)
• If using Node.js < 18, you'll need to install and configure node-fetch

📖 Usage

Basic Import

const { gpt5, deepseek, geminiProxy2, lenna, brave } = require('aryai');

// Or import specific functions
const { gpt5 } = require('aryai');

GPT-5

const { gpt5 } = require('aryai');

async function askGPT5() {
  try {
    const response = await gpt5("What is artificial intelligence?", 1, false);
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

askGPT5();

Parameters:

• prompt (string): Your question or prompt
• uid (string|number, optional): User ID (default: 1)
• reset (boolean, optional): Reset conversation context (default: false)

DeepSeek

const { deepseek } = require('aryai');

async function askDeepSeek() {
  try {
    const response = await deepseek("Explain quantum computing in simple terms");
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

askDeepSeek();

Gemini Proxy (Image Analysis)

const { geminiProxy } = require('aryai');

async function analyzeImage() {
  try {
    const imageUrl = "https://example.com/image.jpg";
    const response = await geminiProxy("What do you see in this image?", imageUrl);
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

analyzeImage();

Gemini Proxy 2 (Text Only)

const { geminiProxy2 } = require('aryai');

async function askGemini() {
  try {
    const response = await geminiProxy2("Write a short story about a robot");
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

askGemini();

GPT-3.5 Turbo

const { gpt35Turbo } = require('aryai');

async function askGPT35() {
  try {
    const response = await gpt35Turbo("What are the benefits of renewable energy?", 1);
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

askGPT35();

Llama-4

const { llama4 } = require('aryai');

async function askLlama() {
  try {
    const response = await llama4("Explain machine learning algorithms", 123);
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

askLlama();

Lenna AI

const { lenna } = require('aryai');

async function askLenna() {
  try {
    const response = await lenna("How does blockchain technology work?");
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

askLenna();

Brave Search

const { brave } = require('aryai');

async function searchBrave() {
  try {
    const response = await brave("latest news about artificial intelligence");
    console.log(response);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

searchBrave();

🎯 Complete Example

const { gpt5, deepseek, geminiProxy2, lenna, brave } = require('aryai');

async function aiComparison() {
  const prompt = "What is the future of artificial intelligence?";
  
  console.log("🤖 Asking multiple AI services the same question...\n");
  
  try {
    // GPT-5 Response
    console.log("🔵 GPT-5 Response:");
    const gpt5Response = await gpt5(prompt);
    console.log(gpt5Response);
    console.log("\n" + "─".repeat(50) + "\n");
    
    // DeepSeek Response
    console.log("🔴 DeepSeek Response:");
    const deepseekResponse = await deepseek(prompt);
    console.log(deepseekResponse);
    console.log("\n" + "─".repeat(50) + "\n");
    
    // Gemini Response
    console.log("🟢 Gemini Response:");
    const geminiResponse = await geminiProxy2(prompt);
    console.log(geminiResponse);
    console.log("\n" + "─".repeat(50) + "\n");
    
    // Lenna Response
    console.log("🟡 Lenna AI Response:");
    const lennaResponse = await lenna(prompt);
    console.log(lennaResponse);
    
  } catch (error) {
    console.error("❌ Error:", error.message);
  }
}

aiComparison();

📝 API Reference

Available Functions

| Function | Description | Parameters | |----------|-------------|------------| | gpt5(prompt, uid, reset) | GPT-5 API wrapper | prompt, uid (optional), reset (optional) | | deepseek(prompt) | DeepSeek API wrapper | prompt | | geminiProxy(prompt, imgUrl) | Gemini image analysis | prompt, imgUrl | | geminiProxy2(prompt) | Gemini text-only | prompt | | gpt35Turbo(prompt, uid, reset) | GPT-3.5 Turbo wrapper | prompt, uid (optional), reset (optional) | | llama4(prompt, uid, url) | Llama-4 Maverick | prompt, uid (optional), url (optional) | | lenna(prompt) | Lenna AI wrapper | prompt | | brave(prompt) | Brave search | prompt |

Error Handling

All functions return promises and should be used with async/await or .then()/.catch() for proper error handling:

// Using async/await (recommended)
try {
  const response = await gpt5("Your prompt here");
  console.log(response);
} catch (error) {
  console.error("API Error:", error.message);
}

// Using .then()/.catch()
gpt5("Your prompt here")
  .then(response => console.log(response))
  .catch(error => console.error("API Error:", error.message));

🔧 TypeScript Support

This package works great with TypeScript. Here's an example:

import { gpt5, deepseek, geminiProxy2 } from 'aryai';

interface AIResponse {
  status: boolean | string;
  message?: string;
  error?: string;
  [key: string]: any;
}

async function getAIResponse(prompt: string): Promise<AIResponse> {
  try {
    const response: AIResponse = await gpt5(prompt, 1, false);
    return response;
  } catch (error) {
    throw new Error(`AI API Error: ${error instanceof Error ? error.message : 'Unknown error'}`);
  }
}

🚀 Deployment on Replit

This package is perfect for Replit projects! Simply install it in your Replit project:

1. Open the Shell in your Replit project
2. Run: npm install aryai
3. Import and use in your code

📊 Rate Limits & Best Practices

• Be mindful of API rate limits
• Implement proper error handling
• Use appropriate uid values for session management
• Cache responses when possible to reduce API calls

🤝 Contributing

Contributions are welcome! Please feel free to submit issues, feature requests, or pull requests.

📄 License

MIT License - see the LICENSE file for details.

🙋‍♂️ Support

If you encounter any issues or have questions:

1. Check the examples above
2. Review the error messages (they're descriptive)
3. Open an issue on the repository

🔗 Related

• Aryan's API Documentation
• Built for Node.js 18+
• Works great with Express.js, Next.js, and other Node.js frameworks

Made with ❤️ by Aryan Chauhan