OpenRouterLite SDK

The OpenRouterLite SDK provides a simple way to interact with an OpenRouter-compatible API without manually writing HTTP requests.

It helps with:

• API authentication
• request formatting
• chat completions
• streaming responses
• cleaner integration in TypeScript and JavaScript projects

Installation

Install the SDK using npm:

npm install openrouterlite-sdk

or using yarn:

yarn add openrouterlite-sdk

Replace openrouterlite-sdk with your final published package name.

Import the SDK

After installation, import the SDK into your project:

import OpenRouterLite from "openrouterlite-sdk";

Initialize the Client

Create a client instance using your API key:

const client = new OpenRouterLite({
  apiKey: process.env.OPENROUTER_API_KEY,
});

You can also pass the API key directly:

const client = new OpenRouterLite({
  apiKey: "YOUR_API_KEY",
});

Basic Chat Completion

Send a simple chat completion request:

const response = await client.chat.completions.create({
  model: "gpt-4",
  messages: [
    {
      role: "user",
      content: "Hello!",
    },
  ],
});

console.log(response);

Example Response

{
  "id": "chatcmpl-123",
  "object": "chat.completion",
  "choices": [
    {
      "message": {
        "role": "assistant",
        "content": "Hello! How can I help you today?"
      }
    }
  ]
}

Sending Multiple Messages

You can send a full conversation by passing multiple messages:

const response = await client.chat.completions.create({
  model: "gpt-4",
  messages: [
    {
      role: "system",
      content: "You are a helpful assistant",
    },
    {
      role: "user",
      content: "Explain artificial intelligence",
    },
  ],
});

console.log(response);

Streaming Responses

Streaming allows you to receive generated output chunk by chunk:

const stream = await client.chat.completions.create({
  model: "gpt-4",
  stream: true,
  messages: [
    {
      role: "user",
      content: "Write a short poem",
    },
  ],
});

for await (const chunk of stream) {
  console.log(chunk);
}

Extracting Streamed Text

To print only the streamed text content:

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

Using a Custom API Endpoint

If you are running your own OpenRouter-compatible backend, you can provide a custom base URL:

const client = new OpenRouterLite({
  apiKey: "YOUR_API_KEY",
});

Error Handling

Wrap API calls in try/catch for safer integrations:

try {
  const response = await client.chat.completions.create({
    model: "gpt-4",
    messages: [
      {
        role: "user",
        content: "Hello",
      },
    ],
  });

  console.log(response);
} catch (error) {
  console.error("API Error:", error);
}

Parameters

| Parameter | Type | Description | | ------------- | --------- | ---------------------------------- | | model | string | Model used to generate completions | | messages | array | Conversation messages | | max_tokens | number | Maximum tokens to generate | | temperature | number | Controls randomness | | stream | boolean | Enables streaming responses |

Full Example

import OpenRouterLite from "@your-username/openrouterlite-sdk";

async function run() {
  const client = new OpenRouterLite({
    apiKey: process.env.OPENROUTER_API_KEY,
  });

  const response = await client.chat.completions.create({
    model: "gpt-4",
    messages: [
      {
        role: "user",
        content: "Explain TypeScript in simple words",
      },
    ],
  });

  console.log(response);
}

run();

Notes

• Make sure your API key is valid before making requests.
• For production apps, store secrets in environment variables instead of hardcoding them.
• If your backend supports streaming, set stream: true to consume partial responses in real time.

License

MIT