Summary

In this chapter, we discussed a variety of topics. We started with LLMs in the computer vision domain: ViT for image classification, DetR for object detection, and SD for text-to-image generation. Next, we learned how to fine-tune an LLM with the Transformers library. Finally, we used LangChain to implement a novel LLM-driven application.

In the next chapter, we’ll depart from our traditional topics and dive into the practical field of MLOps.