Choosing the Right Embedding Model: A Guide for LLM Applications

Ryan Nguyen
18 min readJun 6, 2023

Optimizing LLM Applications with Vector Embeddings, affordable alternatives to OpenAI’s API and how we move from LlamaIndex to Langchain

So you may think that I’m gonna write part 2 of the series on how to build a great chatbot app that is different from 99% of tutorials on the internet. Guess what, it is not gonna happen in this post. I’m sorry in advance but there is a reason why I’m not rushing into part 2 yet and I shall explain to you.

Yep, I know you’ve been eagerly awaiting the second part of our journey in building an amazing AI-powered chatbot using Large Language Models (LLMs). But, you know what? Something happened while I was working on my very own LLM app. I discovered that each embedding model I experimented with produced different and intriguing results. Some were simply so good, while others fell a bit short of expectations. That got me thinking:

how can we truly grasp the power of these embedding models and understand their impact on chatbot performance?

So, I couldn’t resist the urge to share my insights with you through this article. Trust me, it’s well worth your time to…

--

--