hiruthicShaSS/MMMCP

3.2

If you are the rightful owner of MMMCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

This server utilizes both OpenAI's GPT-3.5 and Google's Gemini Pro models to generate text in parallel.

Multi-Model MCP Server

This is an MCP server that generates text using both OpenAI's GPT-3.5 and Google's Gemini Pro models in parallel.

Setup

Install dependencies:

pip install -r requirements.txt

Set environment variables:
1. Set up your API keys:
  - Set OPENAI_API_KEY environment variable for OpenAI
  - Set GOOGLE_API_KEY environment variable for Google Gemini
2. Set models:
  - ACTIVE_PROVIDERS=openai,gemini,nvidia
3. Model Selection
  - OPENAI_MODEL=gpt-3.5-turbo
  - GEMINI_MODEL=gemini-2.0-flash
  - FINAL_MODEL=gemini-2.0-flash

Running the Server

python server.py

Usage

The server exposes a tool called generate_text that takes a prompt and returns responses from both models:

response = await generate_text(prompt="Tell me a joke")
print("OpenAI response:", response["openai"])
print("Gemini response:", response["gemini"])