Sylix Node.js SDK

The official Node.js/TypeScript library for the Sylix AI API, providing convenient access to Charles AI models from Node.js applications.

⚠️ Important: This library is intended for server-side usage only. Using it in client-side browser code will expose your API key.

Installation

npm install sylix

Requirements: Node.js 18+

Usage

import Sylix from 'sylix';

const client = new Sylix({
  apiKey: process.env['SYLIX_API_KEY'], // This is the default and can be omitted
});

async function main() {
  const response = await client.charles.chat.completions.create({
    model: 'charles-s1:latest',
    messages: [{ role: 'user', content: 'Hello, Charles!' }],
  });

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

main();

Streaming

We support streaming responses using Server Sent Events (SSE):

import Sylix from 'sylix';

const client = new Sylix();

async function main() {
  const stream = await client.charles.chat.completions.create({
    model: 'charles-s1:latest',
    messages: [{ role: 'user', content: 'Write a short poem' }],
    stream: true,
  });

  for await (const chunk of stream) {
    process.stdout.write(chunk.choices[0]?.delta?.content || '');
  }
}

main();

Request & Response Types

This library includes TypeScript definitions for all request params and response fields:

import Sylix from 'sylix';
import { ChatRequest, ChatResponse, ChatMessage } from 'sylix';

const client = new Sylix();

const messages: ChatMessage[] = [
  { role: 'system', content: 'You are a helpful assistant.' },
  { role: 'user', content: 'What is TypeScript?' },
];

const params: ChatRequest = {
  model: 'charles-s1:latest',
  messages,
  temperature: 0.7,
};

const response: ChatResponse = await client.charles.chat.completions.create(params);

Available Models

| Model | ID | Description | |-------|-----|-------------| | Charles Mini | charles-mini:latest | Lightweight, fast responses | | Charles S1 | charles-s1:latest | Balanced performance for coding | | Charles V1 | charles-v1:latest | Advanced reasoning capabilities | | Charles 2.9 | charles-2.9:latest | State-of-the-art performance | | Charles R1 | charles-r1:latest | Deep reasoning and analysis |

import { CHARLES_MODELS } from 'sylix';

// Use model constants for type safety
const response = await client.charles.chat.completions.create({
  model: CHARLES_MODELS.S1, // 'charles-s1:latest'
  messages: [{ role: 'user', content: 'Hello!' }],
});

Handling Errors

When the API returns a non-success status code, an error is thrown:

import Sylix, { SylixError } from 'sylix';

const client = new Sylix();

async function main() {
  try {
    const response = await client.charles.chat.completions.create({
      model: 'invalid-model',
      messages: [{ role: 'user', content: 'Hello' }],
    });
  } catch (err) {
    if (err instanceof SylixError) {
      console.log(err.code);    // e.g., 'INVALID_MODEL'
      console.log(err.message); // Detailed error message
    } else {
      throw err;
    }
  }
}

main();

Error Codes

| Code | Description | |------|-------------| | MISSING_API_KEY | API key not provided | | INVALID_MODEL | Model not found or unavailable | | INVALID_API_KEY | Invalid or expired API key | | RATE_LIMIT_ERROR | Too many requests | | STREAM_ERROR | Streaming connection error | | NETWORK_ERROR | Network connectivity issue |

Retries

The library automatically retries on connection errors, 408, 429, and 5XX status codes. You can configure this:

const client = new Sylix({
  maxRetries: 2, // default is 4
  timeout: 30000, // default is 30 seconds
});

Function Calling

const response = await client.charles.chat.completions.create({
  model: 'charles-v1:latest',
  messages: [{ role: 'user', content: 'What is the weather in Tokyo?' }],
  tools: [
    {
      type: 'function',
      function: {
        name: 'get_weather',
        description: 'Get current weather for a location',
        parameters: {
          type: 'object',
          properties: {
            location: { type: 'string', description: 'City name' },
          },
          required: ['location'],
        },
      },
    },
  ],
  tool_choice: 'auto',
});

// Check if the model wants to call a function
if (response.choices[0].message.tool_calls) {
  const toolCall = response.choices[0].message.tool_calls[0];
  console.log('Function:', toolCall.function.name);
  console.log('Arguments:', toolCall.function.arguments);
}

Embeddings

Generate vector embeddings for text:

const response = await client.charles.embeddings.create({
  model: 'charles-mini:latest',
  input: 'The quick brown fox jumps over the lazy dog',
});

console.log(response.data[0].embedding); // Vector array

Models API

// List all available models
const models = await client.charles.models.list();
console.log(models.data);

// Get a specific model
const model = await client.charles.models.retrieve('charles-s1:latest');
console.log(model);

Configuration

import Sylix from 'sylix';

const client = new Sylix({
  apiKey: process.env['SYLIX_API_KEY'], // Required
  baseURL: 'https://api.sylixide.com/v1', // Default
  timeout: 60000, // Request timeout in ms
  maxRetries: 4, // Retry attempts for failed requests
});

Environment Variables

| Variable | Description | |----------|-------------| | SYLIX_API_KEY | Your Sylix API key | | SYLIX_BASE_URL | Override the default base URL |

Semantic Versioning

This package follows SemVer conventions.

We recommend pinning to a specific version and upgrading periodically:

npm install [email protected]

Requirements

• Node.js 18 or higher
• TypeScript 4.7+ (for TypeScript users)

Support

• Documentation: docs.sylixide.com
• API Reference: api.sylixide.com
• Issues: GitHub Issues
• Email: [email protected]

License

Proprietary License © 2026 Sylix Technologies. All rights reserved.

See LICENSE for details.