LLM Dialog

Interactive tool for exploring how Large Language Models (LLMs) converse with each other.

Features

• Pre-built conversation scenarios:
  • Never Say Yes Game
  • AI Ethics Tribunal
  • Improv Comedy Show
  • Agile Methodology Debates
  • Language Translation
  • Philosophy Face-offs
• Custom system prompts for each bot
• Multiple LLM model support including:
  • Llama 3.2 (90b and 11b variants)
  • Gemma 2 9b
  • Mixtral 8x7b
  • OpenAI GPT-4o Mini
  • Claude 3 Haiku
  • Gemini 1.5 Flash models
• Real-time streaming responses
• Editable conversation history

Usage

1. Select two LLM models to converse
2. Set custom instructions for each model
3. Start the conversation with an initial prompt
4. Click "Continue" to let models respond to each other
5. Use pre-built scenarios from the demo cards
6. Edit or delete conversation turns as needed

Setup

Prerequisites

• Modern web browser with ES Modules support
• Web server for local development

Local Setup

1. Clone this repository:

git clone https://github.com/gramener/llmdialog.git
cd llmdialog

2. Serve the files using any static web server. For example, using Python:

python -m http.server

3. Open http://localhost:8000 in your web browser

Deployment

On Cloudflare DNS, proxy CNAME llmdialog.straive.app to gramener.github.io.

On this repository's page settings, set

• Source: Deploy from a branch
• Branch: main
• Folder: /

Technical Details

Architecture

• Frontend: Vanilla JavaScript with lit-html for rendering
• LLM Integration: Multiple model providers through LLM Foundry API
• Styling: Bootstrap 5.3.3 with dark mode support

Dependencies

All dependencies are loaded via CDN:

• lit-html v3 - Template rendering
• Bootstrap v5.3.3 - UI components
• marked v13 - Markdown parsing
• asyncllm v1 - LLM API integration

License

MIT