📖What it means

The unit a language model reads and writes in, and the unit it is billed by. Cost is tokens times the model's rate, which is why sending the same context repeatedly is expensive even when nothing new is being asked.

📖Easily confused with

[{"term": "Cost", "distinction": "Tokens are the quantity; cost is what they came to."}, {"term": "Rate card", "distinction": "Tokens are consumed; the rate card is what each one costs."}]

📖Also called

- model token
- billing unit

📖Filed under

Category: general.

Service: Cost Center.