You can run an AI agent locally on your laptop:

No need to connect or pay a provider like OpenAI or Anthropic.

Combine the open source Google Gemma 3 model, Smolagents and LM Studio and you're ready:


Here are the steps:

1️⃣ Host a Local LLM

Download and install LM Studio to host an AI model locally.
Download a suitable LLM (like Gemma 3 12B).


2️⃣ Set up development environment

Install Python and Visual Studio Code.


3️⃣ Use Smolagent library

Utilize the Smolagent Python library for building AI agents easily.


4️⃣ Generate Agent Code with Local LLM

Use the hosted LLM in LM Studio to generate the Python code for your desired AI agent.

Provide a detailed prompt with instructions, an example, and relevant documentation (like Lite-LLM).


5️⃣ Configure Connection to Local LLM

In the generated Python code, configure it to connect to the locally running LLM in LM Studio by specifying the API base URL (found in LM Studio's developer tab) and the correct model ID.


6️⃣ Run the AI Agent

Execute the Python script in VS Code.

The agent will then interact with the local LLM in LM Studio to perform the defined tasks.


More info on Snapdragon on Windows and the Snapdragon X Elite I use in the video:

https://lnkd.in/eD-Cpuba