Our provider offers a fast tier: output streams four times faster at twice the price. Which of our four AI products should pay for it?
Make your call first. Use the published answer and rubric to examine your reasoning when you are ready.
Case briefFacts and constraints for your answer
All figures are an illustrative planning fixture. A provider's fast tier streams output four times faster at twice the price: 200 tokens a second instead of 50.
| Voice agent | Spoken replies of about 60 tokens. |
|---|---|
| Chat assistant | Answers of about 300 tokens, read as they stream. |
| Coding agent | 40 model calls of 800 output tokens per task; a task costs $1.20 on the standard tier. |
| Nightly summaries | Generated overnight; nobody waits for them. |
Your task. Decide which products should pay for the fast tier, show the waiting time each one removes, and say how you would confirm the decision.
0 words