Are LLMs smarter in some languages than others?
- Track:
- PyData: LLMs
- Type:
- Poster
- Level:
- intermediate
- Room:
- Exhibit Hall
- Start:
- 13:00 on 11 July 2024
- Duration:
- 60 minutes
Abstract
Have you ever asked yourself if Large Language Models (LLMs) perform differently across various languages? I have.
In this poster session, I will demonstrate how tokens, embeddings, and the LLMs themselves perform when utilized in 30 different languages. I will illustrate how languages influence pricing and various model characteristics.
Spoiler:
- The Greek language is the most expensive to process by most models.
- Processing Asian languages on Gemini is cheaper.
- You can save up to 15% of tokens by removing diacritics.
