Inference Cost
The compute cost of an AI model producing an answer on each call; if you use an API, it is usually billed on the tokens processed. In classic software one more customer costs almost nothing, while here every query incurs a real expense.
AI and GrowthTerm 62 / 81
In practice
There is no single "cost per query" figure: providers price input and output separately, and cached input and batch processing as separate line items. Caching has separate prices for writes and reads; if the hit rate is low, the saving is smaller than expected (our reading). If you set price for the average user, cost can exceed revenue on the heavy user; track cost as a distribution, not an average (our reading). Take the current table from the provider's page and do not bake an old figure into your model.