Multi-AI Chat Debate

AI Group Chat is an multi-AI conversation platform designed to facilitate dynamic debates, collaborative coding, and complex discussions between various AI personalities. The platform leverages multiple LLM providers to create a rich, interactive environment where users can watch different models interact, challenge each other, and solve problems in real-time.

Core Features

The application is built with a focus on conversation flow to ensure high-quality interactions.

• Multi-Model Support: Integration with DeepInfra and Fireworks.ai to utilize models like DeepSeek, Qwen, and GLM.
• Dynamic Personalities: Create custom AI participants with specific expertise, communication styles, and system instructions.
• Intelligent Flow Control: A dedicated Flow Service manages turn-taking, prevents repetition, and handles @mentions to maintain natural conversation rhythms.
• Real-time Streaming: Messages are streamed via Socket.io, providing a live, responsive chat experience.
• Database Persistence: All conversations, participants, and messages are stored in a PostgreSQL database.
• Docker Integration: Full containerization support for both the application and the database.

Technical Stack

• Backend: Node.js with Express
• Real-time Communication: Socket.io
• Database: PostgreSQL
• Frontend: Vanilla JavaScript, HTML5, CSS3
• Markdown Rendering: Marked.js with Highlight.js for code blocks
• Containerization: Docker and Docker Compose

Prerequisites

Before starting the installation, ensure you have the following installed on your system:

• Node.js (v20 or higher)
• PostgreSQL (if running locally)
• Docker and Docker Compose (optional, for containerized setup)
• API Keys for DeepInfra or Fireworks.ai

Installation and Setup

Local Development Setup

1. Clone the repository

git clone https://github.com/TonyGeez/debate-ai
cd debate-ai

2. Install dependencies

npm install

3. Configure environment

cp .env.example .env
# Edit .env and add your API keys and database credentials

4. Setup database and start application

# Initialize PostgreSQL database
npm run db:setup

# Start development server
npm run dev

The application will be available at http://localhost:3000

Prerequisites:

• Node.js 20+ installed
• PostgreSQL 15+ running locally
• API keys for DeepInfra and/or Fireworks AI

Docker Deployment

1. Clone the repository

git clone https://github.com/TonyGeez/debate-ai
cd debate-ai

2. Configure environment

cp .env.docker.example .env.docker
# Edit .env.docker and add your API keys

3. Build and start containers

# Build images
docker compose build

# Start services (includes automatic database initialization)
docker compose up -d

The application will be available at http://localhost:3000

To stop services:

docker compose down

Prerequisites:

• Docker and Docker Compose installed
• Ports 3000 and 5432 available

---

Environment Configuration

The application requires several environment variables to function correctly. These should be defined in your .env or .env.docker file.

• PORT: The port the server will listen on (default: 3000).
• DB_HOST: Database host (use db for Docker).
• POSTGRES_USER: PostgreSQL username.
• POSTGRES_PASSWORD: PostgreSQL password.
• POSTGRES_DB: Database name.
• FIREWORKS_API_KEY: Your API key from Fireworks.ai.
• DEEPINFRA_API_KEY: Your API key from DeepInfra.
• DEFAULT_MESSAGE_LIMIT: Default cap on messages per conversation.

Contributing

Contributions are welcome to help improve the AI Group Chat platform. Please feel free to submit pull requests or open issues for bugs and feature requests.

---

Todo

Essential Features

• Delete chat from sidebar
• Edit chat settings (max message, title, topic)
• Rename conversation
• Export conversation (JSON, Markdown, PDF)
• Search conversations by title/topic
• Search within conversation content
• Copy message to clipboard
• Edit sent messages (with edit history)
• Delete individual messages
• Reply to specific message (threaded replies)
• Proper @mention support with autocomplete
• Typing indicators for AI participants
• User authentication system
• Fork/duplicate conversation
• Archive conversations
• Bulk actions (delete/archive multiple)

AI & Conversation Features

• Add/remove AI participants mid-conversation
• AI participant library with pre-made personalities
• Providers managemeng (API models endpoint)
• Custom API endpoints
• AI model switching during conversation
• Conversation analytics (speaker stats, sentiment analysis)
• AI confidence scores visualization
• Advanced flow controls (custom turn-taking logic)
• Conditional AI responses based on triggers
• AI memory management and optimization
• Long-term conversation archiving

Media & Rich Content

• File attachments (PDFs, documents)
• Code execution sandbox
• Real-time collaborative code editing
• Video message support
• File drag-and-drop interface
• Media gallery view per conversation

Advanced Features

• Conversation branching (multiple parallel topics)
• Speech-to-text input
• Text-to-speech for AI responses
• Plugin system for extensibility
• WebRTC voice/video chat
• AI model comparison mode (side-by-side responses)
• A/B testing framework for AI personalities
• Advanced moderation tools
• Spam detection for AI responses
• Conversation templates and presets
• Smart conversation suggestions

DevOps & Performance

• CI/CD pipeline (GitHub Actions)
• Automated E2E testing (Playwright)
• Unit test coverage >80%
• Health check endpoints
• Auto-scaling configuration
• Backup and restore automation
• Caching layer (Redis) for frequently accessed data