An Introduction to the Ray AI Libraries

In the last chapter you’ve learned about Ray Core, and how it can be used to build distributed applications. This chapter introduces you to the main concepts of Ray’s Artificial Intelligence (AI) libraries and how you can utilize them for creating and deploying typical ML workflows. We will demonstrate how to use those libraries by building an application that fine-tunes a large language model (LLM), deploys it for online inference, and uses it for offline batch inference. We will also explain when and why you should use Ray’s AI libraries and provide a brief overview of the Ray AI ecosystem. Lastly, we will delve into the connection between Ray and other systems.