Glossary · AI-native development
Cost per Successful Task
Total system cost divided by the number of tasks that satisfy a defined success criterion, including retries, failed runs, tool use, and evaluation overhead.
Why it matters
A cheap model call can produce an expensive workflow if it fails often or requires repeated human correction.
In practice
Measure provider charges and infrastructure cost across 100 repository tasks, then divide by the number whose patches pass tests and review.
Common confusion
Cost per token measures usage. Cost per successful task measures useful outcomes.
Related terms
Browse the learning paths to see this term in context — every lesson is free to read.