Comment by famouswaffles

There's fixed compute per token but more tokens = more compute so a LLM will technically have more "time" for a query with more tokens preceding it.