roundtable-mcp
Integrates OpenAI's GPT-4o model into multi-model roundtable discussions, enabling it to provide independent responses and critique or synthesize answers from other participating models.
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@roundtable-mcpRun a roundtable debate on whether AI will replace software engineers"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
roundtable-mcp
An MCP server that lets Claude orchestrate multi-model AI roundtable discussions between GPT-4o, Gemini 2.0 Flash, Grok-3, and DeepSeek-R1.
How It Works
Round 1 — Independent Responses: Each model answers the prompt independently, with no awareness of the others.
Round 2 — Cross-Critique: Every model sees all Round 1 responses and provides critique, refinement, or synthesis.
Tools
Tool | Description |
| Full 2-round discussion with configurable presets |
| Single-round quick poll (no cross-critique) |
| List available models and API key status |
| List available discussion presets |
Presets
debate — Structured debate with strong positions
brainstorm — Creative ideation session
analysis — Deep analytical examination
devil — Devil's advocate challenges
consensus — Seek common ground and synthesize
Installation
git clone https://github.com/brandondek/roundtable-mcp.git
cd roundtable-mcp
pip install -e .API Keys
Set the following environment variables for each model you want to use:
OPENAI_API_KEY=sk-... # GPT-4o
GEMINI_API_KEY=AI... # Gemini 2.0 Flash
XAI_API_KEY=xai-... # Grok-3
DEEPSEEK_API_KEY=sk-... # DeepSeek-R1Models without a configured API key will be skipped gracefully.
Claude Desktop Configuration
Add this to your claude_desktop_config.json:
{
"mcpServers": {
"roundtable": {
"command": "roundtable-mcp",
"env": {
"OPENAI_API_KEY": "sk-...",
"GEMINI_API_KEY": "AI...",
"XAI_API_KEY": "xai-...",
"DEEPSEEK_API_KEY": "sk-..."
}
}
}
}Usage Examples
Once installed, ask Claude:
"Run a roundtable debate on whether AI will replace software engineers"
"Quick poll all models: what's the best programming language for beginners?"
"Brainstorm session on novel uses for LLMs in healthcare"
"Devil's advocate discussion on remote work vs office"
License
MIT
This server cannot be deployed
Related MCP Connectors
Multi-model AI debates: GPT-4o, Claude, Gemini & 200+ models discuss, then synthesize insight.
Convene a panel of expert AI personas to debate any decision from every side.
Ask GPT, Claude, Gemini and Grok at once: one consensus answer plus where they agree and split.
mumo is a remote MCP server for multi-model deliberation. Your agent sends a question to models from different labs — Claude, GPT, Gemini, Grok, DeepSeek, Kimi, and more — and gets back their full responses plus typed cross-model reactions. The participating models react to each other directly and explain, in their own words, what they agree with, challenge, or want to explore further. Agents can run a deliberation with `create_deliberation`, wait for results with `wait_for_round`, and steer follow-up rounds with `append_round` using structured signals like KEEP, EXPLORE, CHALLENGE, CORE, and SHIFT. Built for architecture, plan/spec review, strategy, and pre-launch pressure tests. Free tier available. API key required.