jatevo-x402-sdk

Simple axios interceptor for Jatevo x402 LLM API - Works in terminals, VSCode, and Node.js

🆕 Version 1.3.0 - GLM 4.6 Support

Added support for the latest GLM 4.6 model with enhanced reasoning capabilities. Simple axios interceptor pattern that works perfectly in terminals, VSCode, and any Node.js environment!

Features

• 🎯 Simple Integration - Just wrap your axios instance with our interceptor
• 🖥️ Terminal/CLI Ready - Works perfectly in Node.js, terminals, and VSCode
• 🚀 No API Keys Required - Your wallet address is your authentication
• 💰 Pay-Per-Use - Only pay for what you actually use ($0.01 per request)
• ⚡ Instant Settlement - Payments handled through PayAI facilitator
• 🌐 Dual Network Support - Choose between Base or Solana for payments
• 🤖 Multiple Models - Access to Qwen, GLM, Kimi, DeepSeek, and GPT-OSS
• 🔒 Secure - EIP-3009 compliant (Base) and SPL token transfers (Solana)
• 📦 TypeScript Support - Full type definitions included
• 🔑 Private Key Support - Works with private keys for terminal/CLI usage

Installation

npm install jatevo-x402-sdk
# or
yarn add jatevo-x402-sdk
# or
pnpm add jatevo-x402-sdk

Quick Start

const { withPaymentInterceptor } = require('jatevo-x402-sdk');
const axios = require('axios');

// Your EVM private key (keep this secure!)
const PRIVATE_KEY = process.env.PRIVATE_KEY;

// Create axios client with payment interceptor
const client = withPaymentInterceptor(axios.create(), PRIVATE_KEY);

// Make API requests - payments are handled automatically
const response = await client.post('https://api.jatevo.ai/chat/completions/qwen', {
  messages: [
    { role: 'system', content: 'You are a helpful assistant.' },
    { role: 'user', content: 'Explain async/await in JavaScript' }
  ],
  temperature: 0.7,
  max_tokens: 200
});

console.log(response.data.choices[0].message.content);

How It Works

1. Request Sent: Your request is sent to the Jatevo API
2. 402 Response: If payment is required, the API returns a 402 status
3. Auto-Payment: SDK creates a payment header using your private key
4. Retry Request: Request is automatically retried with payment
5. Get Response: You receive the API response seamlessly

Configuration Options

The SDK uses a simple pattern with minimal configuration:

const { withPaymentInterceptor } = require('jatevo-x402-sdk');
const axios = require('axios');

// Basic usage with private key
const client = withPaymentInterceptor(
  axios.create(),  // Your axios instance
  privateKey        // Your EVM private key (for Base) or Solana private key
);

Available Models

| Model ID | Name | Description | Price | |----------|------|-------------|-------| | qwen | Qwen 3 Coder 480B | Advanced code generation and analysis | $0.01 | | glm-4.6 | GLM 4.6 | Latest GLM model with enhanced reasoning | $0.01 | | glm-4.5 | GLM 4.5 | Advanced reasoning and problem-solving | $0.01 | | kimi | Kimi K2 | Long context understanding | $0.01 | | deepseek-r1-0528 | DeepSeek R1 0528 | Latest reasoning model | $0.01 | | deepseek-v3.1 | DeepSeek V3.1 | Efficient chat model | $0.01 | | gpt-oss | GPT-OSS | OpenAI's efficient model | $0.01 |

Example Usage

Basic Chat Completion

const { withPaymentInterceptor } = require('jatevo-x402-sdk');
const axios = require('axios');

async function chat() {
  const client = withPaymentInterceptor(
    axios.create(),
    process.env.PRIVATE_KEY
  );
  
  try {
    const response = await client.post(
      'https://api.jatevo.ai/chat/completions/qwen',
      {
        messages: [
          { role: 'user', content: 'What is the capital of France?' }
        ],
        temperature: 0.7,
        max_tokens: 100
      }
    );
    
    console.log(response.data.choices[0].message.content);
  } catch (error) {
    console.error('Error:', error.message);
  }
}

chat();

Streaming Responses

const response = await client.post(
  'https://api.jatevo.ai/chat/completions/kimi',
  {
    messages: [...],
    stream: true
  },
  {
    responseType: 'stream'
  }
);

response.data.on('data', chunk => {
  console.log(chunk.toString());
});

Using Different Models

// Qwen model
await client.post('https://api.jatevo.ai/chat/completions/qwen', {...});

// GLM models
await client.post('https://api.jatevo.ai/chat/completions/glm-4.6', {...});  // Latest
await client.post('https://api.jatevo.ai/chat/completions/glm-4.5', {...});

// Kimi K2 model
await client.post('https://api.jatevo.ai/chat/completions/kimi', {...});

// DeepSeek models
await client.post('https://api.jatevo.ai/chat/completions/deepseek-r1-0528', {...});
await client.post('https://api.jatevo.ai/chat/completions/deepseek-v3.1', {...});

// GPT-OSS model
await client.post('https://api.jatevo.ai/chat/completions/gpt-oss', {...});

Error Handling

The SDK provides clear error messages:

try {
  const response = await client.post(...);
} catch (error) {
  if (error.response?.status === 402) {
    console.error('Payment required but failed:', error.message);
  } else if (error.response?.status === 429) {
    console.error('Rate limit exceeded');
  } else {
    console.error('API error:', error.message);
  }
}

Network Support

The SDK automatically handles payments on both Base and Solana networks:

• Base Network: Fast settlement (~200ms) using USDC on Base L2
• Solana Network: Reliable settlement (~400ms) using USDC on Solana

Simply provide your private key and the SDK handles the rest - no network configuration needed!

Migration from v1.x

If you were using v1.0.x or v1.1.x, update your code:

Old Pattern (v1.0.x/v1.1.x)

// Complex client initialization
const client = new X402Client({
  baseUrl: 'https://api.jatevo.ai',
  privateKey: PRIVATE_KEY,
  network: 'base'
});

const response = await client.chat.completions(...);

New Pattern (v1.2.0) - Simpler!

// Simple axios interceptor
const client = withPaymentInterceptor(axios.create(), PRIVATE_KEY);
const response = await client.post('https://api.jatevo.ai/chat/completions/qwen', {...});

Benefits of v1.2.0

• Simpler API: Just one function - withPaymentInterceptor
• No Browser Dependencies: Works in any Node.js environment
• Smaller Bundle: Reduced from ~50KB to ~5KB
• Better Error Handling: Clear error messages
• PayAI Integration: Automatic retry on payment failures

TypeScript Types

import type {
  X402Config,
  LLMMessage,
  LLMRequest,
  LLMResponse,
  ModelType,
  ModelInfo,
  PaymentRequirements,
  EVMPaymentData,
  SolanaPaymentData
} from 'jatevo-x402-sdk';

Support

• Documentation: https://jatevo.ai/x402/documentation
• Website: https://jatevo.ai
• Issues: https://github.com/jatevo/x402-sdk/issues

License

MIT © Jatevo