Gemini LLM Chatbot | Gemini Pro Model

The project is a Streamlit-based chatbot application using Google's Gemini LLM (Generative AI) model. Users can input a question, and the chatbot responds in real-time by generating answers from the Gemini model. The app maintains a chat history, which updates with both the user's queries and the AI's responses, allowing for an interactive conversation. It integrates the Google Gemini API to handle AI-powered conversation and displays the dialogue using the Streamlit interface.

Services:
  • LLM
  • Genai
Client:

Personal

Project link:
www.flatheme.net
Duration:

N/A

This chatbot application is built using the Gemini LLM provided by Google. It integrates environmental variables and a Streamlit interface to interact with users. The chatbot not only responds to user queries in real-time but also maintains a history of the conversation, allowing for context-aware interactions in a session.


Features

  • Environment Variables: Manages sensitive data using Python's dotenv package.
  • Streamlit Interface: Offers a user-friendly, web-based interface for real-time interactions.
  • Chat History: Maintains a session-based history of conversations for context retention.
  • Gemini LLM Integration: Leverages Google's Gemini LLM for dynamic and intelligent responses.

Prerequisites

  • Python 3.8+
  • Streamlit
  • dotenv Python package
  • os
  • genai

Usage

  • Enter your query in the text input box and submit.
  • View both your question and the Gemini's response displayed on the screen, with previous interactions saved in session state for context.

© 2024 A S M Morshedul Hoque, All Rights Reserved.