I think both things can be true: new models benchmark higher and eat more tokens.
From my experience new models are slower and use more tokens even on questions which gpt 4 answered correctly. It is mostly because newer models tend to be more verbose (even with prompt requesting short answers).
alt Hacker News
From my experience new models are slower and use more tokens even on questions which gpt 4 answered correctly. It is mostly because newer models tend to be more verbose (even with prompt requesting short answers).