Euri SDK

Node.js SDK for interacting with the Euri Chat Completions API.

Installation

npm install euri

Usage

Basic Usage

import { EuriClient } from 'euri';

const client = new EuriClient({
  apiKey: 'YOUR_API_TOKEN'
});

async function main() {
  try {
    // Quick completion with a simple prompt
    const response = await client.complete('What is AGI?', 'gpt-4.1-nano');
    console.log('Assistant:', response);
  } catch (error) {
    console.error('Error:', error);
  }
}

main();

Chat Completion

import { EuriClient, CoreMessage } from 'euri';

const client = new EuriClient({
  apiKey: 'YOUR_API_TOKEN'
});

async function main() {
  try {
    const messages: CoreMessage[] = [
      {
        role: 'user',
        content: 'What is AGI?'
      },
      {
        role: 'assistant',
        content: [
          {
            type: 'text',
            text: 'AGI stands for Artificial General Intelligence. It refers to....'
          }
        ]
      },
      {
        role: 'user',
        content: 'I didn\'t get it.'
      }
    ];

    const response = await client.createChatCompletion({
      model: 'gpt-4.1-nano',
      messages: messages,
      max_tokens: 1000,
      temperature: 0.7
    });

    console.log('Assistant:', response.choices[0].message.content);
  } catch (error) {
    console.error('Error:', error);
  }
}

main();

Available Models

The Euri API supports the following models:

• llama-3.3-70b-versatile (default)
• gemini-2.0-flash-001
• mistral-saba-24b
• llama-4-scout-17b-16e-instruct
• llama-4-maverick-17b-128e-instruct
• gemini-2.5-pro-exp
• gpt-4.1-nano
• gpt-4.1-mini
• deepseek-r1-distill-llama-70b (with reasoning extraction)
• qwen-qwq-32b (with reasoning extraction)

API Reference

EuriClient(config)

Creates a new client instance.

• config.apiKey (string, required): Your Euri API key.
• config.baseURL (string, optional): Override the default API base URL.

client.createChatCompletion(request)

Creates a chat completion based on the provided messages and model.

• request.model (string): The model to use
• request.messages (array): Array of message objects
• request.max_tokens (number, optional): Maximum tokens to generate (default: 1000)
• request.temperature (number, optional): Sampling temperature (default: 0.7)

Returns a promise resolving to the chat completion response.

client.complete(prompt, model, options)

Convenience method for simple completions with a single prompt.

• prompt (string): The user prompt
• model (string, optional): The model to use (default: 'llama-3.3-70b-versatile')
• options (object, optional):
  • max_tokens (number): Maximum tokens to generate
  • temperature (number): Sampling temperature

Returns a promise resolving to the text response as a string.

Development

1. Clone the repository
2. Install dependencies: npm install
3. Build the package: npm run build

License

MIT