smollm2-1.7b-runner

A Node.js package to run SmolLM locally using node-llama-cpp. This package enables text generation using a lightweight LLM model.

Model

This package uses the SmolLM2-1.7B-Instruct model from Hugging Face.

Installation

npm install smollm2-1.7b-runner

Usage

const smol = require("smollm2-1.7b-runner");

async function chat() {
  const response = await smol.prompt("Hello, how are you!");
  console.log(response);
}

chat();

Functions

prompt()

smol.prompt(userPrompt, options);

• userPrompt (string): The input text to generate a response from the model.
• options (object, optional):
  • maxTokens (number): Maximum number of tokens to generate (default: 200). Range: 1-2048.
  • temperature (number): Controls randomness in output (default: 0.8). Range: 0.1-1.0.
• Returns: Generated text response from the model.
• Throws: Error if initialization failed or invalid options are provided.