What are output tokens?
Output tokens are what the model writes back, and they are the expensive half.
~/ai-glossary
$ glossary
11 of 14 in #the-basics
They cost several times what input costs, because generating text takes far more work than reading it.
This is the one to watch. Asking for a summary instead of a rewrite, or a diff instead of the whole file, is the difference between hitting your limit at lunchtime and not hitting it at all.
Was this page helpful?