The answer
Emojis are represented by multiple unicode bytes, which often get split into multiple tokens by the tokenizer.
What is Tokenization and How do LLMs Read Text?
Why do emojis consume more tokens?
More FAQs in What is Tokenization and How do LLMs Read Text?
Subwords allow the model to handle typos, prefixes, and suffixes efficiently without expanding the vocabulary index into millions of words.
LLM providers charge per token processed, not per word or character. Long prompts cost more because they translate to more tokens.
A word or character the tokenizer doesn't recognize, which is replaced by a special [UNK] token.
Still have questions?
Browse all our FAQs or reach out to our support team
Want to upskill yourself?
Your learning library
Small steps. Big possibilities.
Find your next course. Build something great.