Skip to content

Getting Started

This section describes the steps you need to take to build and test your own intelligent agents on your own local computer. The ideal configuration is to have a local GPU with at least 12GB of RAM on the GPU. Although you can still run agents on a local computer without a GPU, the response times might be so slow that you will become frustrated. My local GPU is a NVIDIA RTX 2080 Ti which returns 52 tokens per second.

You can test the tokens per second on your local computer using our lab on the Deep Learning Course.

If you are not an expert on setting up Python, then setting up your local system can be VERY frustrating.

Picking the Right Agent Library

Picking the Right Library

Installing GPU Drivers

GPU Driver Version

Conda Setup

Conda Setup

Testing Python Libraries

Testing Libraries

Ollama Setup

Ollama Setup