Kronos Labs JavaScript/TypeScript Client

Official JavaScript/TypeScript client for the Kronos Labs API.

Installation

npm install kronoslabs
# or
yarn add kronoslabs

Quick Start

JavaScript (CommonJS)

const { KronosLabs } = require("kronoslabs");

// Initialize the client
const client = new KronosLabs({ apiKey: "your-api-key-here" });

// Non-streaming chat completion
async function chat() {
  const response = await client.chat.completions.create({
    prompt: "Hello, how are you?",
    model: "hyperion", // or "hermes"
    temperature: 0.7,
    isStream: false,
  });

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

chat();

TypeScript

import { KronosLabs } from "kronoslabs";

const client = new KronosLabs({ apiKey: "your-api-key-here" });

async function chat() {
  const response = await client.chat.completions.create({
    prompt: "Hello, how are you?",
    model: "hyperion", // or "hermes"
    temperature: 0.7,
    isStream: false,
  });

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

chat();

Streaming

const client = new KronosLabs({ apiKey: "your-api-key-here" });

async function streamChat() {
  const stream = await client.chat.completions.create({
    prompt: "Tell me a story",
    model: "hyperion", // or "hermes"
    temperature: 0.8,
    isStream: true,
  });

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

streamChat();

Features

• Simple and intuitive API similar to OpenAI's ChatGPT client
• Full TypeScript support with type definitions
• Support for multiple models: hyperion and hermes
• Support for both streaming and non-streaming responses
• Automatic handling of API authentication
• Comprehensive error handling
• Works in Node.js environments
• Tool calling (coming soon)

API Reference

Initialize Client

const client = new KronosLabs({
  apiKey: "your-api-key",
  baseUrl: "https://kronoslabs.org", // optional
});

Chat Completions

const response = await client.chat.completions.create({
  messages: [], // Optional: array of message objects
  prompt: "Your prompt here", // Required: the prompt text
  model: "hermes", // Optional: "hyperion" or "hermes" (default: "hyperion")
  temperature: 0.7, // Optional: controls randomness (0.0-2.0)
  isStream: false, // Optional: enable streaming
  tool: false, // Optional: enable tool usage (Work in Progress - not yet functional)
});

Parameters

• messages (array, optional): Array of message objects with role and content properties
• prompt (string, required): The prompt text to send to the model
• model (string, optional): Model to use - either "hyperion" or "hermes". Default: "hyperion"
• temperature (number, optional): Controls randomness in the response (0.0-2.0). Default: 0.7
• isStream (boolean, optional): Enable streaming responses. Default: false
• tool (boolean, optional): Work in Progress - Tool calling functionality is currently under development and not yet available. Default: false

Response Format

Non-streaming response:

{
  id: string;
  object: 'chat.completion';
  created: number;
  model: string;
  choices: [{
    index: number;
    message: {
      role: 'assistant';
      content: string;
    };
    finish_reason: string;
  }];
  usage: {
    prompt_tokens: number;
    completion_tokens: number;
    total_tokens: number;
  };
}

Streaming response (async iterable of chunks):

{
  id: string;
  object: 'chat.completion.chunk';
  created: number;
  model: string;
  choices: [{
    index: number;
    delta: {
      role?: 'assistant';
      content?: string;
    };
    finish_reason: string | null;
  }];
}

Error Handling

import { KronosLabs, KronosLabsError } from "kronoslabs";

const client = new KronosLabs({ apiKey: "your-api-key" });

try {
  const response = await client.chat.completions.create({
    prompt: "Hello",
  });
} catch (error) {
  if (error instanceof KronosLabsError) {
    console.error("API error:", error.message);
    console.error("Status code:", error.statusCode);
  }
}

Examples

See the examples directory for more usage examples:

• basic-chat.js - Simple non-streaming chat
• streaming-chat.js - Streaming responses
• chat-with-history.js - Maintaining conversation history
• error-handling.js - Handling errors

License

MIT License