HALTGPT

Advanced OpenAI Chat Interface

A feature-rich desktop application that provides an enhanced interface for interacting with OpenAI's language models. Featuring multiple chat tabs, agent-based interactions, custom instructions, and various tools to improve your AI experience.

Python OpenAI API DALL-E Multi-Agent PyQt5
Download View Source

This application is an independent project and is not developed, endorsed, or officially associated with OpenAI. This application uses the OpenAI API as a third-party service.

Key Features

📑

Multiple Chat Pages

Maintain separate conversations in up to 20 tabs, making it easy to organize different topics and contexts.

🤖

Agent Mode

Enable single agent or multi-agent dialogs for specialized interactions with different roles and personalities.

🎭

Custom Instructions

Set system and developer instructions for consistent AI behavior across all your conversations.

🎨

Image Generation

Generate images with DALL-E 3 and DALL-E 2 models directly within the application interface.

🎯

Custom Actions

Create and manage reusable prompt templates and actions to streamline your workflow.

💾

Session Management

Save and load your conversation sessions for continued work across different times.

System Requirements

💻

Operating System

Windows 10/11, macOS 11+, or Linux (Ubuntu 20.04+)

🔑

API Key

OpenAI API key with ChatGPT access (all models supported)

📶

Connectivity

Internet connection for API calls (offline mode for cached responses)

Model Selection

Available Models

Choose from various OpenAI models to suit your needs:

  • GPT-4o
  • GPT-4-turbo
  • GPT-4
  • GPT-3.5-turbo
  • And more specialized models

Image Generation

Create images with OpenAI's DALL-E models:

  • DALL-E 3
  • DALL-E 2
  • Multiple size options

Model Parameters

Fine-tune your AI responses with customizable parameters:

Temperature: Control creativity level (0.0-1.0)
Top_p: Nucleus sampling parameter
Max Tokens: Control response length

Installation

# Clone the repository
git clone https://github.com/Sevsai/HALTGPT.git
cd HALTGPT

# Create a virtual environment
python -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Install dependencies
pip install -r requirements.txt

# Set up your API key
echo "OPENAI_API_KEY=your_api_key_here" > .env

# Run the application
python .venv/HALTGPT.py

Quick Start

  1. Get an OpenAI API Key:
    • Visit OpenAI's Platform
    • Sign up or log in to your account
    • Navigate to the API keys section and create a new key
    • Store the key in the .env file as shown above
  2. Launch the Application:
    • Run python .venv/HALTGPT.py from the command line
    • The interface will display with chat tabs on the left and settings on the right

Using HALTGPT

Basic Chat

Getting started with HALTGPT is simple:

  1. Enter your prompt in the input area of any chat tab
  2. Click "Generate" or press Ctrl+Enter to get a response
  3. View the response in the output area
  4. Use the chat tabs at the top to manage multiple conversations

Agent Mode

Create advanced AI interactions using Agent Mode:

  1. Go to the "Agents" tab in the right panel
  2. Enable "Agent Mode" with the checkbox
  3. Configure agent name and instructions
  4. For multi-agent dialog, enable both "Agent Mode" and "Multi-Agent Dialog"
  5. Configure the number of agents and their roles
  6. Use the chat to generate conversations between the agents

Image Generation

Create AI-generated images directly in the app:

  1. Go to the "Tools" tab
  2. Enter a prompt in the "Image Generation" section
  3. Select the model (DALL-E 3 or DALL-E 2) and size
  4. Click "Generate Image" to create an image
  5. Use "Save Image" to save the generated image

Keyboard Shortcuts

Ctrl+Enter
Generate response for current tab
Ctrl+Tab
Switch to next tab
Ctrl+Shift+Tab
Switch to previous tab
Ctrl+S
Save output to file
Ctrl+L
Clear current chat history
Esc
Stop generation

Custom Actions

Create shortcuts for your most common tasks:

  1. Click "Manage Actions" in any chat tab
  2. Create custom actions to insert text templates or execute operations
  3. Use these actions from the dropdown menu in chat tabs

Customization

Model Settings

Configure temperature, top_p, and max tokens in the "Chat" tab of the settings panel.

Instructions

Set system and developer instructions in the "Instructions" tab to guide the AI's behavior across all chats.

Appearance

Customize the application's look in the "Appearance" tab:

  • Choose from Light, Dark, and Blue themes
  • Adjust font size and family
  • Create custom themes

Troubleshooting

API Authentication Issues

+

If you encounter "Authentication Error" messages, verify that your API key is correct and has not expired. You can check the API key status in your OpenAI account dashboard. If issues persist, try regenerating your API key.

High Token Usage

+

If you're seeing unexpected token consumption, adjust the "Context Retention" setting in the preferences menu to reduce the amount of conversation history sent with each request. You can also enable "Intelligent Summarization" to automatically compress long conversations.

Document Import Problems

+

For issues with document imports, ensure you have the necessary dependencies installed by running pip install -r requirements-extras.txt. For large documents, try using the "Chunk Import" option which breaks the document into smaller pieces for processing.

Performance Optimization

+

If the application feels sluggish, try enabling "Lightweight Mode" in settings, which reduces background processing. For large conversation histories, use the "Archive Old Conversations" feature to improve performance while maintaining searchability.