(Fake) LLM ChatBot with FastApi (Backend) and Streamlit (Frontend)

This project demonstrates how to create a real-time conversational AI by streaming responses from an LLM. It uses FastAPI to create a web server that accepts user inputs and streams generated responses back to the user and streamlit for creating a simple chatbot interface.

Getting Started

Prerequisites

• Python 3.11
• Poetry -> pip install poetry
• Docker (optional, for containerized deployment) [Container have been tested in Window environment, for Mac might be needed some adjustment]
• Make (to use Makefile commands)

Running Application on local machine

1. Install dependencies:

   make install_env

2. Run Backend Application:

   make run_backend

3. Run Frontend Application:

   make run_backend

Running Application on docker containers

1. Build images

   make docker_build_app

2. Run containers

   make docker_app_up

   Note: In the logs, it may appear that the app is deployed on 0.0.0.0:, but it is actually accessible on localhost:.