How to Run Multi-LLM Agents on Raspberry Pi with CrewAI


Running multiple Large Language Model (LLM) agents on a Raspberry Pi might sound like a hefty task, but with the introduction of CrewAI, orchestrating role-playing, autonomous AI agents for collaborative intelligence just got easier and more accessible. This blog will guide you through setting up CrewAI on a Raspberry Pi to harness the power of multi-agent systems for various tasks, from trip planning to stock analysis.

Why CrewAI?

CrewAI stands out by enabling AI agents to assume roles, share goals, and operate as a cohesive unit, paving the way for sophisticated multi-agent interactions. This framework is ideal for projects requiring collaborative intelligence, such as smart assistant platforms, automated customer service, or multi-agent research teams.

Getting Started on Raspberry Pi

Before diving into the setup, ensure your Raspberry Pi is running a compatible OS and has Python installed. CrewAI and its dependencies are Python-based, making it suitable for Raspberry Pi's ARM architecture.

  1. Installation

First, install CrewAI and any additional packages required for your agents to function. For instance, if your agents will perform web searches, you might need the duckduckgo-search package.

pip install crewai
pip install duckduckgo-search
  1. Setting Up Your Crew

After installing the necessary packages, you can start defining your agents, their roles, goals, and the tasks they will perform. Here's an example of setting up a crew with a researcher and a writer agent:

import os
from crewai import Agent, Task, Crew

os.environ["OPENAI_API_KEY"] = "YOUR_API_KEY"

researcher = Agent(
  role='Senior Research Analyst',
  goal='Uncover cutting-edge developments in AI',

writer = Agent(
  role='Tech Content Strategist',
  goal='Craft compelling content on tech advancements',

task1 = Task(description="Analyze latest AI advancements", agent=researcher)
task2 = Task(description="Develop an engaging blog post", agent=writer)

crew = Crew(agents=[researcher, writer], tasks=[task1, task2])
result = crew.kickoff()


Running on Raspberry Pi

Given the Raspberry Pi's hardware constraints, it's essential to manage resources wisely. CrewAI, with its efficient design, allows for running lightweight agents without overloading the system. However, when connecting your crew to models, especially large LLMs, consider using remote connections to models hosted on more powerful servers or cloud platforms, rather than running them directly on the Raspberry Pi.

Examples and Applications

CrewAI's versatility opens up a myriad of applications, from generating landing pages to analyzing stocks or planning trips. Each task can be tailored to the specific goals and roles of the agents involved, demonstrating the framework's adaptability to different scenarios.

How CrewAI Compares

CrewAI's unique approach to orchestrating AI agents sets it apart from other frameworks by emphasizing role-based agent design, autonomous inter-agent delegation, and flexible task management. Its compatibility with open-source models further enhances its applicability in various settings, from hobbyist projects to research and development.

Contribution and Support

CrewAI's open-source nature encourages contributions from the community. Whether you're improving the framework, adding new features, or providing feedback, your input is valuable for its growth and evolution.


Setting up CrewAI on a Raspberry Pi showcases the potential of running sophisticated multi-agent systems on compact and affordable hardware. By following the steps outlined in this guide, you can embark on creating intelligent, collaborative agent-based applications, pushing the boundaries of what's possible with Raspberry Pi and AI.


To further enhance your journey with CrewAI on Raspberry Pi, we've compiled a list of essential resources. These links will provide you with additional information, tools, and examples to explore and utilize as you integrate CrewAI into your projects.

  • CrewAI Official Website: Dive deeper into the capabilities of CrewAI and discover how it can transform your Raspberry Pi into a powerful AI hub. Visit for more information, documentation, and updates.

  • CrewAI GitHub Repository: Access the source code, contribute, and explore the development progress of CrewAI on GitHub. Check out the repository at for a closer look at the codebase and to contribute to the project.

  • CrewAI Examples: For practical applications and inspiration on what you can achieve with CrewAI, visit the examples repository. It's a treasure trove of real-world use cases and sample projects. Explore these examples at to kickstart your own AI projects.

These resources are designed to support you at every step of your CrewAI journey, from learning the basics to deploying sophisticated multi-agent systems on your Raspberry Pi. Whether you're looking for technical documentation, seeking to contribute to the project, or searching for inspiration through examples, you'll find everything you need to make the most out of CrewAI.