ijyliu / anlp23-project

An empirical study of the costs and practicalities of prompt engineering techniques on standard and novel benchmarks
0 stars 0 forks source link

Gains per cost in cents, not tokens #70

Closed ijyliu closed 9 months ago

ijyliu commented 9 months ago

relative costs can differ some for GPT-4 since there are different input/output prices

ijyliu commented 9 months ago

possibly cut average length vs direct prompting figure and replace with this