Skip to content

froma-dev/local-webllm

Repository files navigation

lightning

WebLLM Chat Bot

Overview

The WebLLM Chat Bot is a simple web application that utilizes the WebLLM library to provide a conversational AI experience.
Users can input prompts, and the bot will respond with relevant and helpful text.
The bot is designed to be a basic, chat-like interface that showcases the capabilities of the WebLLM technology.

More info about web-llm: https://github.com/mlc-ai/web-llm and https://webllm.mlc.ai/

Features

  • Users can input prompts in a text field
  • The bot responds with relevant and helpful text
  • Responses are displayed in a chat-like section
  • No authentication or account creation required
  • Clear chat context

Technical Details

  • Built using HTML, CSS, and Vanilla JavaScript
  • Utilizes the WebLLM library for natural language processing and text generation

Getting Started

  1. Clone the project repository to your local machine
  2. Install dependencies npm install
  3. Run npm run dev and open the URL in the browser
  4. Wait for the model to load into the GPU before you can start interaction
  5. Start interacting with the bot by typing in the input field
  6. View the bot's responses in the chat-like section

Limitations:

  • The bot is designed for basic, conversational purposes and may not be suitable for complex or nuanced discussions
  • The bot's responses are generated based on the input prompts, and may not always be 100% accurate or helpful
  • The project is a proof-of-concept and is not intended for production use without further development and testing

Contributing:

  • Contributions are welcome! If you'd like to improve or expand the project, feel free to fork the repository and submit a pull request

About

Chat bot using webLLM running in the browser using the GPU

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published