Which parts of a request count toward token cost?

Question

Accepted Answer

Both input and output — your prompt and the model's reply are each counted — Both the input prompt and the output reply are counted toward token cost. Neither is free, and billing isn't by request count.

Answer

Only the input — just your prompt is billed, and the reply is always free

Answer

Only the output — the model's reply is billed, and your prompt is the free part

Answer

Neither directly — billing is by the request count, not by any token totals

Which parts of a request count toward token cost?

Why this is the answer

More Building with LLMs: APIs, Tokens & Cost flashcards