A team needs to sort incoming customer support tickets into one of three categories, billing, technical, or general, at high volume and low latency, with cost per request kept as low as possible. The task involves no images, no long documents, and no multi-step reasoning, just a short label decision per ticket. Which type of model from the Foundry catalog best fits this task?
Select an answer to reveal the explanation.
Short Explanation
Sorting a ticket into one of three buckets is not a hard problem, so the model doing it shouldn't be doing more work than it needs to. Reaching for the biggest, most capable model in the catalog for a job this simple is like hiring a specialist consultant to sort your mail, you'll pay a lot more and wait longer for the exact same result a much lighter tool could deliver instantly. Paying extra for a model that can also read images doesn't help either, since nothing in this job involves a picture, that's a cost with no payoff today. And a model tuned to write with flair and personality is aimed at the opposite goal from what's needed here, which is a short, consistent, predictable label with no creative flourish. The right call is a smaller, efficient model that matches the actual difficulty of the task, keeping both the response time and the bill down.
Full Explanation
The correct answer is D. A smaller, lower-cost model is well suited to a simple three-way classification task, and since the requirement explicitly prioritizes high volume, low latency, and low cost over anything else, matching model size to task complexity is the right tradeoff. Option A is incorrect because the largest flagship reasoning model is built for complex multi-step tasks; using it here would add unnecessary latency and cost without meaningfully improving accuracy on a task this simple. Option B is incorrect because the scenario states tickets contain no images today, so paying for multimodal capability is a speculative cost for a capability not currently needed; the requirement can be revisited if the input format actually changes. Option C is incorrect because a model tuned for creative writing tone optimizes for stylistic, expressive output, which is the opposite of what a consistent, terse classification label requires.