/ChatGPT-GUI

This program is a chatbot that uses the OpenAI API to generate responses to user queries. It has a GUI with a text box for the user to input their query and a label to display the chatbot's response. The user can also set their OpenAI API key and clear the chatbot's previous response.

Primary LanguagePython

Local LLM Chat GUI

A simple Python-based GUI application that allows you to interact with local Large Language Models through Ollama. This application provides a chat interface with multiple tabs for different conversations.

Features

  • Multiple chat tabs support
  • Local LLM integration via Ollama
  • Simple and intuitive interface
  • Real-time responses
  • Error handling and status messages

Prerequisites

Before using this application, you need to:

  1. Install Ollama on your system

    • Visit Ollama's website to download and install
    • Make sure the Ollama service is running

  2. Install Python dependencies:

    pip install -r requirements.txt

Installation

  1. Clone the repository:

    git clone [your-repository-url]
cd ChatGPT-GUI

  2. Install dependencies:

    pip install -r requirements.txt

Usage

  1. Start Ollama service on your system

  2. Run the application:

    python chatGPT.py

  3. Using the Interface:

    • Click "New Chat" to open a new chat tab
    • Type your message and press Enter to send
    • Use "Delete Chat" to remove the current tab
    • The default model is "llama2" but can be modified in the code

Technical Details

The application is built using:

  • Python 3.x
  • Tkinter for GUI
  • Ollama for local LLM integration
  • Asynchronous request handling

Project Structure

ChatGPT-GUI/
├── chatGPT.py       # Main GUI application
├── requestLocal.py  # Ollama integration
└── requirements.txt # Python dependencies

Error Handling

The application handles common errors including:

  • Ollama service not running
  • Model not found/not downloaded
  • Invalid inputs
  • Connection issues

Troubleshooting

  1. If you get "Ollama server is not running" error:

    • Check if Ollama is installed
    • Verify Ollama service is running
    • Restart Ollama service

  2. If model responses are slow:

    • Check your system resources
    • Consider using a lighter model

Contributing

Feel free to submit issues and enhancement requests!