Intelligent Multi-Agent System with Ollama, MCP, Gemma

Ever dreamed of automating Airbnb bookings, getting real-time stock insights, and solving complex problems—all with AI? 🤖✨ In this step-by-step tutorial, I'll show you exactly how to build a modular, intelligent multi-agent system using Python, MCP tools, Ollama, Gemma 3:4b, Praison AI, DuckDuckGo, and Gemini 2.x Thinking!

You’ll learn:

• How to set up your environment and AI models easily ✅
• Creating specialized agents for Airbnb, Finance, Web Search, and more! 🌐📈
• Intelligent routing and clean user interfaces with Gradio 💻
• Advanced features like context retention and URL filtering 🔥

Watch the video below:

This repository demonstrates an intelligent multi-agent system built with Python, Gradio, praisonaiagents, and MCP tools. The system integrates multiple specialized AI agents—Airbnb bookings, financial data analysis, web search, a local LLM agent, and a sequential thinking agent.

Features

• Airbnb booking agent using MCP.
• Finance agent retrieving real-time data with yfinance.
• Web search agent leveraging DuckDuckGo.
• Local LLM agent using ollama/gemma3:4b.
• Sequential thinking agent with Gemini 2.5 Pro (Gemini 2.5 Pro or previous model).
• Context handling and URL filtering.

Installation

1. Clone the repository:

   git clone YOUR_REPO_URL
   cd YOUR_REPO_DIRECTORY