Recent publication

Comparing Large Language Models

How to decide which is the best state-of-the-art model?

We are excited to share with you our latest research Comparing Large Language Models where we delve into some of the methods used to compare state-of-the-art Large Language Models (LLMs).

Here are the key highlights:

  • Evaluation Methods: In this research, we report some of the methods used to compare state-of-the-art Large Language Models (LLMs).
  • Chatbot Arena: One of the most popular methods is the Chatbot Arena; a leaderboard for the best performing LLMs, obtained via a crowdsourced approach.
  • Quality Trends: Analyzing the Chatbot Arena over time, we show there are indications that the difference in quality between new state-of-the-art LLMs is diminishing.

We believe this analysis provides valuable insights into the evolving landscape of LLMs and their performance metrics.

Request product details

Help & Support

Already a customer?

Office locations

Contact LSEG near you