saphan cost
Shows what a stream has cost, from the run record: the number of runs, each run with its model, its cost in US dollars, its status and its token counts (input, cached, output and reasoning), and the s
Shows what a stream has cost, from the run record: the number of runs, each run with its model, its cost in US dollars, its status and its token counts (input, cached, output and reasoning), and the stream's total. Token subtotals are given per backend and model only, because different vendors count tokens differently; no cross-vendor token total is printed. A run whose transcript holds no readable usage says so instead of showing a number, and costs that are not tied to a run (such as refusals) are named as not attributable to the stream rather than estimated.
Global options apply to every command — see Every command.