Comment by danenania
1 day ago
Model routing is deceptively hard though. It has halting problem characteristics: often only the smartest model is smart enough to accurately determine a task's difficulty. And if you need the smartest model to reliably classify the prompt, it's cheaper to just let it handle the prompt directly.
This is why model pickers persist despite no one liking them.
Yes but prompt evaluation is far faster than inference as it can be done (mostly) in parallel, so I don't think that's true.
The problem is that input token cost dominates output token cost for the majority of tasks.
Once you've given the model your prompt and are reading the first output token for classification, you've already paid most of the cost of just prompting it directly.
That said, there could definitely be exceptions for short prompts where output costs dominate input costs. But these aren't usually the interesting use cases.
No, you're talking about costs to user, which are oversimplifications of the costs that providers bear. One output token with a million input tokens is incredibly cheap for providers
3 replies →
That's usually not the case for thinking models. And usually hard problems have a very short prompt.
2 replies →
but if the less strong model has low false positives you can just route them in order of strength
That's a very big "if".