Skip to content

Saga-Labs/foro

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

6 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

AI Consensus

A simple web app where you can ask a question to multiple AI models and watch them discuss it together to reach a consensus answer.

What is it?

Ask any question, and 3 AI models will:

  1. Share their initial perspectives
  2. Discuss and build on each other's ideas
  3. Work together to find common ground
  4. Present a consensus summary

You can see their full thought process and watch them collaborate in real-time.

Features

  • Multiple AI Models: Configure 3 different models from various providers
  • Visible Discussion: Watch models discuss in a familiar chat interface
  • Thought Process: Optionally view each model's reasoning before their response
  • Consensus Summary: Get a final summary highlighting what they agreed on
  • Flexible Setup: Mix and match models from different providers

Supported AI Providers

  • OpenAI (GPT-4, GPT-3.5, etc.)
  • Anthropic (Claude models)
  • Google (Gemini models)
  • OpenRouter (Access to 100+ models)
  • Local Models (Ollama, LM Studio)

Quick Start

1. Download

git clone https://github.com/Saga-Labs/foro.git
cd foro

2. Run

python3 -m http.server 8000

Then open http://localhost:8000 in your browser

3. Configure Models

  1. Click the ⚙️ Settings button
  2. For each of the 3 models:
    • Choose a provider
    • Select a model
    • Enter your API key (if required)
  3. Save settings

4. Ask a Question

Type any question you'd like the models to discuss and reach consensus on, then click send.

Getting API Keys

Cloud Providers:

Local Models (No API Key Needed):

Example Questions

  • What's the best way to learn programming?
  • How will AI impact education in the next decade?
  • Should we prioritize space exploration or ocean research?
  • What are the most important skills for future careers?
  • How can we make cities more sustainable?

Privacy

Your API keys are stored locally in your browser only. They are never sent anywhere except to the AI providers you choose.

How It Works

  1. Round 1: Each model shares its initial perspective
  2. Round 2+: Models acknowledge each other's points and build toward consensus
  3. Summary: One model generates a consensus summary of the discussion

You can configure the number of discussion rounds in settings (1-5 rounds).

Tips

  • Use different types of models for more diverse discussions
  • Enable "Show thought" to see each model's reasoning process
  • Try 2-3 rounds for most questions
  • Pause/Stop the discussion at any time

Troubleshooting

  • ✅ Make sure you entered valid API keys
  • ✅ Check your internet connection
  • ✅ For local models, ensure the server is running
  • ✅ Open browser console (F12) to see error messages

About

Watch two AI agents have intelligent conversations! Choose from different conversation modes and see AI agents discuss, debate, brainstorm, or make predictions on any topic you give them.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

 
 
 

Contributors