💡 How to use these prompts:
Replace everything in [BRACKETS] with your specific details before sending.
Click Copy to copy any prompt to your clipboard instantly.
Take this sample of my real text and estimate the token count under each model family I name below. Then recalculate the cost per million for each at the same amount of actual text rather than the same token count. Show both numbers.
Read this pricing page and list every threshold where the rate changes: context length tiers, volume tiers, effort levels, peak and off-peak windows. For each one, state whether crossing it rebills the whole request or only the excess. Quote the exact wording.
Given my workload description below, estimate what proportion of my input tokens would be cache reads rather than fresh input. Recalculate the monthly cost at that cache ratio for each model, and show how the ranking changes between zero and eighty percent cache hits.
Given these two models with these completion rates on my kind of work, calculate the cost per SUCCESSFULLY COMPLETED task rather than per call. Assume a failed attempt costs full price and produces nothing. Show how many retries flip the ranking.
Here is a description of a typical day of my usage. Build me a line-by-line estimate of the daily bill for each model, then multiply to a month. Flag any assumption you had to make rather than burying it.